AN ABSTRACT OF THE DISSERTATION OF

Cheng-Yao Chen for the degree of Doctor of Philosophy in Molecular and Cellular Biology presented on December 9, 2004. Title: Biochemical Characterization and Mutational Analysis of Human Uracil-DNA Glycosylase.

Abstract approved: Redacted for privacy Michael I. Schimerlik

PCR-based codon-specific random mutagenesis and site-specific mutagenesis were performed to construct a library of 18 amino acid changes at Arg276 in the conserved leucine-loop of the core catalytic domain of human uracil-DNA glycosylase (UNG). Each Arg276 mutant was then overproduced in E. coli cells and purified to apparent homogeneity by conventional chromatography. All of the R276 mutant proteins formed a stable complex with the uracil-DNA glycosylase inhibitor protein (Ugi) in vitro, suggesting that the active site structure of the mutant enzymes was not perturbed. The catalytic activity of all mutant proteins was reduced; the least active mutant, R276E, exhibited 0.6% of wild-type UNG activity, whereas the most active mutant, R276H, exhibited 43%. Equilibrium binding measurements utilizing a 2- aminopurine-deoxypseudouridine DNA substrate showed that all mutant proteins displayed greatly reduced base flipping/DNA binding. However, the efficiency of UV- catalyzed cross-linking of the R276 mutants to single-stranded DNA was much less compromised. Using a concatemeric [32PJUA DNA polynucleotide substrate to assess enzyme processivity, UNG was shown to use a processive search mechanism to locate successive uracil residues, and Arg276 mutations did not alter this attribute. A transient kinetics approach was used to study six different amino acid substitutions at Arg276 (R276C, R276E, R276H, R276L, R276W, and R276Y). When reacted with double-stranded uracil-DNA, these mutations resulted in a significant reduction in the rate of base flipping and enzyme conformational change, and in catalytic activity. However, these mutational effects were not observed when the mutant proteins were reacted with single-stranded uracil-DNA. Thus, mutations at Arg276 effectively transformed the enzyme into a single-strand-specific uracil-DNA glycosylase. The nuclear form of human uracil-DNA glycosylase (LTNG2) was overproduced in E. coli cells and purified to apparent homogeneity. While UNG2 retained -9 % of 11MG activity, it did form a stable complex with Ugi. Paradoxically, low concentrations of NaC1 and MgC12 stimulated UNG2 catalytic activity as well as the rate of rapid fluorescence quenching; however, the rate of uracil flipping was reduced. When UNG2 bound pseudouracil-containing DNA, conformational change was not detected. Biochemical Characterization and Mutational Analysis of Human Uracil-DNA Glycosylase

by

Cheng-Yao Chen

A DISSERTATION

submitted to

Oregon State University

in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

Presented December 9, 2004 Commencement June 2005 Doctor of Philosotthy dissertation of Cheng-Yao Chen presented on December 9, 2004.

APPROVED:

Redacted for privacy Major Professor, representing Molecular and Cellular Biology

Redacted for privacy

and Cellular Biology Program

Redacted for privacy

Dean of ti 'Graduate School

I understand that my dissertation will become part of the permanent collection of Oregon State University libraries. My signature below authorizes release of my thesis to any reader upon request. Redacted for privacy

Author ACKNOWLEDGMENTS

I would like to thank Drs. Samuel Bennett and Michael Schimerlik for their generous support and expert guidance. Without their assistance, it would have been difficult for me to complete my studies after the tragic death of my Research Advisor, Dr. Dale W. Mosbaugh, on February 17, 2004. It has been a great honor for me to work with such dedicated and knowledgeable scientists. I also acknowledge Drs. William Baird, Jerry Heidel, Victor Hsu, and Tory Hagen for their support and encouragement. I am particularly indebted to my parents, Yao-Nan Chen and Shuei- Yin Chang, for their love and encouragement. Lastly, I would like to express my gratitude to Shu-Ching Hsu for her companionship, thoughtfulness, and optimism. TABLE OF CONTENTS

Page

1 INTRODUCTION...... 1

1.1 Uracil Residues in DNA...... 1

1.1.1 Incorporation of dUMP during DNA Synthesis...... 1 1.1.2 Introduction of Uracil by Cytosine Deamination ...... 3 1.1.3 Biological Consequences of Uracil Residues in DNA...... 5

1.2 Uracil-DNA Glycosylase...... 7

1.2.1 Escherichia coil Uracil-DNA Glycosylase...... 8 1.2.2 Escherichia co/i Double-stranded Uracil-DNA Glycosylase...... 12 1.2.3 Mammalian Uracil-DNA glycosylase...... 15

1.2.3.1 Nuclear and Mitochondrial Uracil-DNA Glycosylase...... 15 1.2.3.2 Human Nuclear and Mitochondrial Form Uracil-DNA Glycosylases and Their Common Catalytic Domain...... 16

1.2.4 Crystal Structures of Uracil-DNA Glycosylases...... 20 1.2.5 Mechanism of Action...... 23

1.3 Other Enzymes with Uracil-DNA Glycosylase Activity in Human Cells... 37

1.3.1 Thymine-DNA Glycosylase...... 3/ 1.3.2 Single-Stranded Selective Monofunctional Uracil-DNA Glycosylase...... 39 1.3.3 Methyl-CpG-Binding Protein 4...... 40

1.4 Uracil-DNA Glycosylase Inhibitor Protein...... 41

1.4.1Bacterial PBS 1/2 Uracil-DNA Glycosylase Inhibitor...... 41 1.4.2 Mechanism of Uracil-DNA Glycosylase Inhibitor Action...... 43

1.5 Uracil-Initiated Base Excision Repair Pathways in Eukaryotes...... 46

1.6 Research Objectives...... 56

2 MATERIALS AND EXPERIMENTAL PROCEDURES...... 57

2.1 Materials...... 57

2.1.1 Chemicals...... 57 TABLE OF CONTENTS (Continued)

Page

2.1.2Radioisotopes...... 58 2.1.3Bacterial Media...... 58 2.1.4Bacterial Strains...... 58 2.1.5Plasmids and Bacteriophage...... 59 2.1.6Chromatographic Resins...... 60 2.1.7Oligonucleotides...... 60 2.1.8Enzymes...... 62 2.1.9Animals...... 63

2.2 Experimental Procedures...... 72

2.2.1 Preparation of Chromatographic Resins...... 72

2.1.1.1 Preparation of DE-52, P11, Sephadex G-75, Bio-Gel P4, and Hydroxyapatite Bio-Gel HTP Resins...... 72 2.2.1.2 Preparation of Dowex 1-X8 Ion Exchange Resins...... 72 2.2.1.3 Preparation of Single-stranded DNA Agarose...... 73

2.2.2 Miscellaneous Methods...... 73

2.2.2.1 Preparation of Dialysis Tubing...... 73 2.2.2.2 Preparation of Oligonucleotides...... 74 2.2.2.3 5'-End Phosphorylation of Oligonucleotides...... 75 2.2.2.4 Annealing Reaction of Duplex DNA...... 75 2.2.2.5 Protein Concentration Measurements...... 75 2.2.2.6 Rapid Protein Staining of Nondenaturing Polyacrylamide Gels...... 76 2.2.2.7 Preparation of Competent Cells...... 76 2.2.2.8 Transfection of Competent E. coli Cells by Electroporation....77

2.2.3 Electrophoresis...... 77

2.2.3.1 Sodium Dodecyl Sulfate Polyacriamide Slab Gel Electrophoresis...... 77 2.2.3.2 Nondenaturing Polyaciylamide Slab gel Electrophoresis...... 78 2.2.3.3 Urea-polyacrylamide Sequencing Gel Electrophoresis...... 79 2.2.3.4 Agarose Gel Electrophoresis...... 79

2.2.4 Purification ofEscherjchja coliDouble-stranded Uracil-DNA Glycosylase...... 80

2.2.5 Western Blot Analysis...... 82 TABLE OF CONTENTS (Continued)

Page

2.2.6 Purification of Polyclonal Antibody forEscherichia coliDouble- stranded Uracil-DNA Glycosylase...... 83

2.2.6.1 Polyclonal Antiserum Production...... 83 2.2.6.2 Dug-sepharose Chromatography...... 84 2.2.6.3 Polyclonal Antibody Purification...... 84

2.2.7Construction of Human UNG* and UNG Overexpression Plasmids...... 93 2.2.8Purification of Core Catalytic Domain of Human Uracil-DNA Glycosylase...... 93 2.2.9 Random and Site-directed Mutagenesis of Core Catalytic Domain of Human Uracil-DNA Glycosylase...... 95 2.2.10 Isolation of R276 Mutants...... 96 2.2.11 Purification of His6-tagged Core Catalytic Domain of Human Uracil-DNA Glycosylase and R276X Mutant Proteins...... 97 2.2.12 Enzyme Assays...... 98

2.2.12.1 Uracil-DNA Glycosylase Activity Assay...... 98 2.2.12.2 Double-strand Uracil-DNA Glycosylase Activity Assay...... 104 2.2.12.3 Nuclease Assay...... 105 2.2.12.4 TJNG'Ugi Binding Assay...... 105

2.2.13UV-Catalyzed Photocrosslinking Reaction...... 106 2.2.14Generation of Uracil-containing Concatemeric DNA Substrate...... 107 2.2.15Restriction Endonuclease Digestion of the Uracil-containing Concatemeric Polynucleotide Substrate...... 108 2.2.16Uracil-DNA Glycosylase Processivity Assay...... 108 2.2.17Steady-state Fluorescence Measurement...... 109 2.2.18Dissociation Constants for DNA Binding...... 110 2.2.19Pre-steady State fluorescence Measurement...... 110 2.2.20Data Analysis...... 111 2.2.21Uracil-DNA Glycosylase Activity Assay (Oligonucleotide Assay)...... 112 2.2.22Purification of Human Uracil-DNA Glycosylase, Nuclear Form....113 2.2.23Matrix-assisted Laser Desorptionllonization Mass Spectrometric Analysis ...... 115

3 Mutational Analysis of Arginine 276 in the Leucine-loop of Human Uracil-DNA Glycosylase (I)...... 116

3.1 Results...... 119 TABLE OF CONTENTS (Continued)

Page

3.1.1 Overproduction and Purification of UNG and Arg276 MutantProteins...... 119 3.1.2 Ability of R276X Mutant Proteins to Bind Ugi ...... 119 3.1.3 Uracil-DNA Glycosylase Activity of R276 Mutants ...... 120 3.1.4 Effect of R276 Mutations on 2-Aminopurine Fluorescent Intensity...... 121 3.1.5 Photochemical Crosslinking of UNG and R276 Mutant Proteins to Single-stranded dU-25-mer...... 131 3.1.6 Processivity of UNG on Concatemeric Uracil-containing [32P]DNA...... 132 3.1.7 Processivity of R276 Mutants on Concatemeric Uracil-containing [32P]DNA...... 139

3.2 Discussion...... 145

4 Mutational Analysis of Arginine 276 in the Leucine-loop of Human Uracil-DNA Glycosylase (II)...... 149

4.1 Results...... 149

4.1.1Fluorescence Properties of dNJU2AP-25-mer...... 149 4.1.2Effect of Arg276 Mutations on UNG-DNA Interactions...... 151 4.1.3Pre-steady State Kinetic Analysis of the UNG Fluorescence Change Associated with Binding Duplex Uracil-DNA...... 152 4.1.4Effect of NaC1 on UNG Fluorescence Quenching Associated with Binding Duplex Uracil-DNA...... 154 4.1.5Effect of Arg276 Mutations on UNG Fluorescence Change Associated with Binding Duplex Uracil-DNA...... 163 4.1.6Pre-steady State Kinetic Analysis of the UNG Fluorescence Change Associated with Binding Single-stranded Uracil-DNA...... 163 4.1.7Effect of Mutations at Arg276 on Uracil-Excision Activity...... 164

4.2. Discussion...... 166

5 Biochemical Characterization of Nuclear Form Human Uracil-DNA Glycosylase: Effects of NaCl or MgC12 on DNA Binding, Uracil Flipping, and Enzyme Conformational Change...... 182

5.1 Results...... 182

5.1.1Overproduction and Purification of Recombinant UNG2...... 182 5.1.2 Ability of UNG2 to Bind Ugi...... 183 TABLE OF CONTENTS (Continued)

Page

5.1.3The Relative Uracil-DNA Glycosylase Activities of UNG2, UNG* and UNG...... 184 5.1.4NaC1 and MgC12 Effects on UNG2 Activity...... 184 5.1.5Effect of NaC1 and MgC12 on UNG2-Binding to Duplex ijiU-DNA...... 185 5.1.6Pre-steady State Kinetics of UNG2-binding to dsU2AP-25-mer 186 5.1.7Effect of NaC1 or MgC12 on UNG2 DNA Binding and Uracil Flipping...... 186 5.1.8Pre-steady State Kinetics of UNG2 Intrinsic Fluorescence Change Induced by Binding to Uracil-DNA...... 187 5.1.9Effect of NaCl and MgC12 on UNG2 Intrinsic Fluorescence Change upon Binding Duplex Uracil-DNA...... 188 5.1.10 Direct Comparison of Time courses of 2AP Fluorescence Enhancement and Intrinsic Protein Fluorescence Change Using a Duplex Uracil-DNA Substrate...... 189

5.2 Discussion...... 190

6 Comparison of the Nucleotide Flipping Mechanism ofEscherichia coliand Human Uracil-DNA Glycosylase...... 224

6.1. Results...... 215

6.1.1Fluorescence Properties of 2AP-containing Oligonucleotides...... 215 6.1.2Pre-steady State Kinetic Analysis of Uracil Flipping by UNG...... 215 6.1.3Direct Comparison of 2AP and Intrinsic Protein Fluorescence Change When Ung or UNG Binds a Double-stranded Uracil- containing DNA...... 216 6.1.4Direct Comparison of UNG and E. coli Ung 2AP Fluorescence Enhancement and Intrinsic Protein Fluorescence Change Using dU2AP-25mer DNA...... 217 6.1.5Direct Comparison ofE. coliUng and UNG Intrinsic Fluorescence Change Induced by Binding to Single-stranded Uracil-containing DNA...... 218 6.1.6Direct Comparison ofE. coliUng and UNG Intrinsic Protein Fluorescence Change Using a single-stranded VU-DNA...... 219

6.2. Discussion...... 232

BIBLIOGRAPHY...... 235 LIST OF FIGURES

Figure Page

1. Scheme of the reaction mechanism for cleavage of the Ni-Cl' glycosylic bondbyE. co/jUng...... 30

2. Interactions leading to stabilization of the oxacarbenium ion-uracil anion intermediate in the E. co/i Ung active site...... 32

3. Deoxyuridine in DNA is severely distorted by the UNG* active center to achieve the observed conformation of the U...... 35

4. Validation ofE.co/i CY01 and CY1 1 strains by cC-excision activity assay ...... 66

5. Validation of E. co/i CY1 1 by PCR of E. co/i ung and dug genes...... 68

6. Validation ofE.co/i CY1O strain lacking recA gene by ultraviolet light exposure ...... 70

7. SDS-polyacrylamide gel analysis of recombinant Dug isolated at various steps during the purification...... 86

8. Western Blot analysis of affinity and specificity of anti-Dug antiserum and purified anti-Dug antibody for Dug protein...... 89

9. Purification of anti-Dug antibody by Dug-sepharose affinity chromatography...... 91

10.SDS-polyacrylamide gel analysis of recombinant UNG* isolated at various steps during the purification...... 100

11.Determination of the polypeptide molecular weight of UNG* by gel filtration chromatography...... 102

12.Tertiary structure of human uracil-DNA glycosylase bound to DNA...... 117

13.Purity of the enzymes used in this study...... 123

14.Ability of UNG and Arg276 mutant proteins to bind Ugi ...... 125

15.Specific uracil-DNA glycosylase activity of R276X mutant proteins...... 127

16.Effect of Arg276 mutations on DNA-binding and base-flipping as measured by 2-aminopurine fluorescence...... 129 LIST OF FIGURES (Continued)

Figure Page

17.Ability of UNG and Arg276 mutant proteins to form UV-catalyzed cross-links to [32P}25-mer DNA...... 133

18.Analysis of reaction products generated by UNG from a concatenated uracil-containing [32P}DNA substrate...... 140

19.Analysis of reaction products generated by Ung, UNG*, IING or its Arg276 mutant proteins from concatenated uracil-containing [32P]DNA... 143

20.Fluorescence properties of 2AP-containing oligonucleotide...... 155

21. DNA binding affinity of UNG and R276X mutants for dijiU2AP-25- mer...... 157 22. A global conformational change of UNG* on binding iiU-DNA...... 159

23.Stopped-flow time trace of UNG intrinsic protein fluorescence change upon binding double-stranded DNA...... 161

24.Tryptophan residues of human uracil-DNA glycosylase...... 170

25.Effect of NaC1 concentration on dsUA-25-mer-induced IJNG intrinsic fluorescence...... 172

26.Stopped-flow time trace of R276 mutant protein intrinsic fluorescence change upon binding dsUA-25-mer...... 174

27.Effect of single-stranded U-25-mer DNA on the intrinsic protein fluorescence of UNG and R276 mutant proteins...... 176

28.Observed rate constants for ssU-25-mer- and dsUA-25-mer-induced intrinsic protein fluorescence change...... 178

29.Uracil base excision activity of UNG and R276 mutant proteins on single- or double-stranded uracil-containing oligonucleotide DNA...... 180

30.Purity of purified recombinant UNG2...... 194

31.NaC1 and MgC12 effects on recombinant UNG2 activity...... 196

32.NaC1 and MgC12 effects on UNG2 for binding to dU-containing double-stranded DNA...... 198 LIST OF FIGURES (Continued)

Figure Page

33.Pre-steady state kinetics of UNG2-induced fluorescence changes of 2AP containing oligonucleotide...... 200

34.Effect of MgC12 on the observed rate of UNG2- and UNG-induced 2AP fluorescence enhancement...... 202

35.Stopped-flow time trace of UNG2 intrinsic protein fluorescence upon binding duplex uracil-DNA...... 204

36.The effect of NaC1 on the transient change in UNG2 intrinsic fluorescence associated with binding dsUA-25-mer...... 206

37.Effect of MgC12 on the transient change in UNG2 intrinsic fluorescence associated with binding dsUA-25-mer...... 208

38.Observed rate constants for the dsUA-25-mer-induced intrinsic protein fluorescence change of UNG2 or UNG in the presence of MgC12...... 210

39.Direct comparison of stopped-flow time traces of 2AP fluorescence enhancement and intrinsic protein fluorescence quenching using dsU'A-25-mer and ds-U'A-25-mer...... 218

40.Fluorescence properties of 2AP-containing oligonucleotides...... 220

41.Pre-steady state kinetic analysis of uracil flipping induced by UNG binding...... 222

42.Comparison of TJNG 2AP fluorescence enhancement and intrinsic protein fluorescence change withE.co/i Ung using duplex uracil-DNA...... 224

43.Comparison of 2AP fluorescence enhancement and intrinsic protein fluorescence quenching induced by Ung and TJNG binding to duplex PU-DNA...... 226

44.Comparison of the change in intrinsic protein fluorescence ofE.co/i Ung and UNG associated with binding single-stranded uracil-DNA...... 228

45.Comparison of the change in intrinsic protein fluorescence of Ung and UNG associated with binding single-stranded VU-DNA...... 230 LIST OF TABLES

Table Page

1. Classification of human DNA polymerases...... 55

2. E. coli strains and genotypes...... 63

3. Uracil-DNA glycosylase activity of JM1O5 transductants...... 64

4. M13 uracil-DNA phage infection assay of E. coli JM1O5 and CY strains 65

5. Buffer conditions for Dug-sepharose affinity chromatography...... 88

6. Purification of catalytic domain of human uracil-DNA glycosylase...... 99 DEDICATION

This dissertation is dedicated to the memory of Dr. Dale W. Mosbaugh, deceased February 17, 2004. Biochemical Characterization and Mutational Analysis of Human Uradil-DNA Glycosylase

1. INTRODUCTION

1.1. Uradil Residues in DNA

1.1.1. incorporation of dUMP during DNA Synthesis

The fidelity of DNA replication and repair synthesis is critically dependent on the synthesis and turnover of deoxynucleoside triphosphate(dNTP) pools,since dNTPsare the immediate substrates for the DNA Polymerases (1). In vivo, low levels of deoxyuridine triphosphate(dUTP)are normally incorporated into DNA in a number of biological systems, including polyoma virus (2), Escherichia coli (3), yeasts (4), and mammalian cells (5,6). Incorporation ofdUTPinto DNA occurs because both prokaryotic and eukaryotic DNA polymerases efficiently utilizedUTPin place of dTTP as a precursor for DNA replication (7-10). As a result, the replacement ofdUTP for dTTP produces U'A rather than TA base pairs. The ability of DNA polymerases to incorporatedUTP intonewly synthesized DNA suggests thatdUTPand dTTP are not easily differentiated by the enzyme when associated with the primer/template. Indeed, the Km values of E. coli DNA polymerase I for incorporation of dTTP and dUTP,4.1 and 5.6 j.tM, respectively, are quite similar (9). Likewise, the Km values for incorporation of dTTP anddUTPby mammalian DNA polymerase a, 1, and y are not dramatically different (8,11,12). Therefore, the frequency ofdUMPincorporation by DNA polymerases is mainly dependent on the relative intracellular pooi size ofdUTP versus dTTP (9). The relative intracellular pool size ofdUTPversus dTTP is controlled by the de novo synthesis ofdUTPin the cell. In E. coli, the de novo synthesis ofdUTP pool is generated both fromdUDPand from dCTP. In the first pathway, the conversion of UDP to dUDP is followed by phosphorylation to yield dUTP. In the other pathway, dCTP is directly converted to dUTP through the action of dCTP deaminase (13-15).

UnlikeE.coil cells, mammalian cells lack dCTP deaminase. Thus, a direct source for the biosynthesis of dUTP is provided by dCMP deaminase, which deaminates dCMP to produce dTJMP (2,16); dUTP then can be synthesized by sequential phosphorylations of dUMP. Alternatively, dUTP can be synthesized by converting rTJDP to dUDP by ribonucleotide diphosphate reductase (17) and subsequently by phosphorylation of dUDP. Because of the differences in the biosynthetic pathways, the intracellular concentration of dUTP inE.coil (0.5 jiM) is significantly greater than that (0.3jiM) in human cells (3,5). However, wild-typeE.coil and mammalian cells do not accumulate dUTP, because of the activity of deoxyuridine 5-triphosphate nucleotidohydrolase (dUTPase), which catalyzes the hydrolysis of dUTP to dTJMP and pyrophosphate (18,19). The reaction catalyzed by dUTPase performs two critical functions: 1) reducing the cellular dUTP concentration and, thus lowering the level of dUTP available to DNA polymerases; 2) providing the substrate dUMP for thymidylate synthase for the synthesis of dTMP (20). Thus,E.coil mutants defective in dUTPase exhibited a dramatically increase in dUTP pool size and a high level of uracil incorporation into the DNA (2 1,22). InE.coli, since dUTP biosynthesis is an obligatory intermediate in the de novo synthesis of dTTP, incorporation of dUMP into

E.coil genome is unavoidable. Although mammalian cells do not maintain as large as a dUTP pool asE.coii, dUTP remains available for DNA synthesis. Measurements of the number of uracil residues per cellular genome have appeared to vary from 1,000 to 498,000 (23-25). These variations may be due to the differences in the biological samples as well as the detection methods, which include GC/Mass spectrometry, HPLC with Till detection, anion-exclusion chromatography, and single-cell gel electrophoresis. In summary, incorporation of dUTP during DNA synthesis inE. coil and mammalian cells occurs naturally and cannot be avoided. 3

1.1.2. Introduction of Uracil by Cytosine Deamination

Uracil can also be introduced into the genome by spontaneous deamination of cytosine residues in DNA and results in the formation of UG mispairs from the pre- existing CG base pairs (2 6,27). The inherent instability of cytosine was first reported by Shapiro and Klein, who suggested that deamination of cytosine, could be involved in spontaneous mutagenesis (28). Two alternative mechanisms for the hydrolytic deamination of cytosine in aqueous solution have been proposed (28). 1) In a direct mechanism, the protonation of theN3position of cytosine is followed by a water molecule or a hydroxyl ion attack at theC4position of the protonated pyrimidine ring, followed by the release of ammonia. 2) The alternative mechanism involves an addition-elimination reaction in which a water molecule is added to the 5,6-double bond of protonated cytosine to form an unstable 5,6-dihydrocytosine residue as an intermediate. Subsequently, a second water molecule attack at the C4 position of the pyrimidine ring results in the loss of ammonia group to yield a uracil hydrate which is followed by dehydration to yield a uracil residue in the fmal step. Under physiological conditions, these reactions may proceed spontaneously or be facilitated by various types of chemical and environmental agents, such as sodium bisulfite and nitrous acids. Under acidic conditions (pH 5.6), high concentration (2.26 M) of sodium bisulfite has been shown to catalyze a time-dependent first order conversion of poly(dC) to poly(dU) nucleotides (29). The reaction proceeds by an acid-catalyzed addition-elimination reaction with the formation of an isomeric 5 ,6-dihydrocytosine-6- sulfonate intermediate. This intermediate is unstable, and hydrolytic deamination of C4 amino group yields 5,6-dihydrouracil-6-sulfonate. In the last step, elimination of bisulfite yields uracil (30). Furthermore, studies have shown that sodium bisulfite- catalyzed deamination of cytosine residues occurs preferentially in single-stranded DNA (29,30). Unlike sodium bisulfite, nitrous acid-promoted deamination of cytosine is relatively nonspecific since it also catalyzes the deamination of adenine and guanine (30). Moreover, deamination of cytosine by nitrous acid in double stranded DNA is as efficient as in single-stranded DNA (30). The rates of hydrolytic deamination of cytosine were initially determined by direct measurements of uracil produced in DNA following incubation either at high temperatures or at extreme pH conditions (31,32). The rate of spontaneous deamination of cytosine in single-stranded DNA, poly(dC), in vitro at 95°C, extrapolated to 37 °C, resulted in a rate constant k = -2x101°residues per second (31). Later, by using a sensitive genetic reversion assay, Frederico and co-workers (33) showed that the rates of spontaneous deamination of cytosine residues in DNA under physiological conditions (37 °C, pH 7.4) were 7x1013and 1 x 10b0 per second for double- and single-stranded DNA, respectively. Thus, the duplex DNA is much more resistant to deamination by a factor of150 as compared to the single-stranded DNA. Since the rate of deamination is more than 150 times greater for single-stranded DNA, it appeared that the processes of replication, recombination, and transcription that involve transient localized denaturation of DNA could accelerate cytosine deamination (32). Finally, by using Frederico's deamination rate constants, the half- time for cytosine deamination in single-and double-stranded DNA in vivo are estimated to be -200 years and30,000 years, respectively. The latter value translates into100 cytosine deamination events per human DNA genome per day (34). The formation of uracil residues in DNA through cytosine deamination can also result from direct photolysis of DNA bases upon ultraviolet (UV) irradiation at wavelengths < 300 nm (35,36). UV irradiation may induce in many cytosine photoproducts including monomeric forms of cytosine hydrates and cytosine glycols (37,38), and dimeric forms such as cyclobutane dimers (CcT, T<>C, C'C) and pyrimidine-pyrimidone (6-4) photoproducts (3 8,39), while cytosine hydrates are the major monomeric photoproduct in DNA (40). Previously, it was demonstrated that persistent cytosine hydrates in irradiated poly(dG)/poly(dC) resulted in the formation of uracil hydrate (41), and further elimination of water molecules from uracil hydrates ultimately generated uracil residues in DNA (42). It has also been demonstrated that 5 irradiation of poly(dC)/poly(dI) nucleotides at 280 nm resulted in the formation of cytosine dimers that further deaminated to uracil dimers with a half-life of 2 hour at 37

°C (43). In addition, using a Ml3mp2Cl4l double-stranded DNA exposed to 160JIm2 UV (254 nm) at 37 °C and pH 7.4, the rate of cytosine deamination in CcC cyclobutane dimers was estimated to be 1.5x106sec, corresponding to a half-life of 5 days at physiological temperature and pH (44). This result is twenty-four times the value (5 h) reported by Barak and coworkers, who used a forward assay to measure the rate of cytosine deamination at multiple cytosine-containing pyrimidine dimer sites in TJV-irradiated plasmid DNA at 37 °C (45). Nonetheless, deamination of cytosine in pyrimidine dimers can readily take place with a half-life of hours or days, rather than years as with spontaneous deamination (33).

1.1.3. Biological Consequences of Uracil Residues in DNA

The importance of excluding uracil residues in DNA was first observed from in vivo studies of E. coli dut strains (3,9,22). These mutants were identified as sof (dnaS) mutants because they showed a higher than normal frequency of recombination and accumulated abnormally short Okazaki-like DNA fragments (9,46). E. coli mutants defective in dUTPase (dut) showed elevated levels of intracellular dUTP pool size and increased incorporation of uracil residues into chromosomal DNA (21,22). It is possible that misincorporation of uracil residues into chromosomal DNA and their subsequent removal from DNA could lead to the production of abnormally short Okazaki-like DNA fragments (3). In addition, the temperature-sensitive dut-1 strain replaces -18 % of thymine with uracil residues in DNA at 42 °C, but displays lethality at the permissive temperature (30 °C) that appears to be associated with the generalized degradation of newly synthesized uracil-containing DNA, 4-S DNA fragments (9). Studies utilizing E. coli ung-1 mutant deficient in uracil-DNA glycosylase and an E. coli dut-1 ung-1 double mutant have led to the understanding of the mechanism for the degradation of uracil-DNA. E. coli dut-1 ung-1 double mutants ri

were found to continue to incorporate dUMP into DNA, but did not generate DNA fragments (3,22). In addition, these double mutants grew normally with a regeneration time that was slightly longer than that of wild-type cells (22). Similar results were observed in bacteriophage containing uracil-DNA. Studies have shown that bacteriophage T4 that had incorporated uracil residues in DNA was unable to establish

a productive infection inE.coli dut mutants possibly due to degradation of uracil- DNA (27,47,48). However, when bacteriophage T4 are grown in E. coli dut ung mutants, the progeny phages incorporate as much as 30 % of uracil residues in place of thymine in DNA, and they are viable and propagate normally (48). These results suggest that the lethal effect associated with high levels of uracil residues in DNA of wild-type cells is a result of uracil-DNA degradation initiated by uracil-DNA glycosylase. Indeed, the adverse effects of misincorporation of dUMP in eukaryotic cells has also been observed. Using the cell-free extracts of human lymphocytes to study DNA synthesis in vitro, it was found the dUTP interfered with normal DNA synthesis, as a majority of small DNA fragments were generated (6). The size of the small DNA fragments depended on the amount of free dUTP added to the extract. Small DNA fragments diminished upon addition of high concentration of free uracil (6 mM), which is an inhibitor of uracil-DNA glycosylase (6). Therefore, misincorporation of dUMP in DNA may have cytotoxic and possibly lethal consequences for cells. Misincorporation of dUMP in DNA has also been found to interfere with protein-DNA interactions. The influence of dUMP incorporation on protein-DNA interactions was first observed in measuring the binding affinity of lac repressor to lac operator inE.coli. Substitution of uracil for thymine at position of 13 of the lac operator DNA sequence resulted in a 10-fold decrease in stability of the repressor- operator complex (49). In addition, it has been reported that the specific binding of nuclear protein(s) from HeLa cell nuclear extract to the cAMP responsive element sequence was reduced when uracil was substitute with thymine (50). Similarly, the incorporation of dUMP into herpes simplex virus type 1 (HSV-1) origin of replication (Ori) interferes with the interaction between Ori and the origin binding protein (OBP) in vitro (50). Moreover, the incorporation of uracil residues into restriction endonuclease recognition sites affects the rates of Hpal, Hindu, and HindlII endonucleases cleavage at corresponding restriction sites (51). All these results suggest that the methyl group of thymine that projects into the major groove of duplex DNA is important for protein-DNA interactions. The generation of uracil residues in DNA through cytosine deamination results in UG mispairs in place of normal U'A base pairs. If these pre-existing UG mispairs in DNA are not repaired prior to DNA replication, they can lead to GC to AT transition mutations (26,27,52). Several studies have suggested that cytosine deamination events in DNA contribute to the overall spontaneous mutation frequency of the cell. Analysis of spontaneous mutation spectra in an E. coli lad gene that was cloned into a derivative of bacteriophage M13 showed that 93 % of the observed mutations occurred through GC to A'T transitions (53). Similarly, analysis of spontaneous mutation spectra at the adenine phosphoribosyl transferase gene locus of Chinese hamster ovary (CHO) cells also revealed a predominant GC to AT transition mutation (81 %) (54). E. coli ung mutant cells were shown to exhibit a -P30 fold increase in G.0 to AT transition mutations as compared to that of ung proficient cells (27). In addition, a 20-fold increase in spontaneous mutation frequency in the target SUP4-o tRNA gene was observed in a Saccharomyces cerevisiae strain deleted in the UNG1 gene that encoded uracil-DNA glycosylase. However, this mutator effect was totally suppressed in the ungi deficient strain carrying the wild-type UNG1 gene on a multicopy plasmid (55). Thus, uracil-DNA glycosylase plays an important role in suppressing G'C to A'T transition mutations due to the cytosine deamination in DNA.

1.2. Uradil-DNA Glycosylase

Uracil-DNA glycosylase catalyzes the hydrolysis of the Ni-Cl' glycosylic bond linking the uracil base to the deoxyribose phosphate backbone of DNA and produces two reaction products: the free uracil base and abasic site-containing DNA (56). Removal of the uracil base constitutes the first step in the multi-step DNA repair pathway known as base excision repair (BER) (57). The enzyme removes uracil residues in single and/or double-stranded DNA arising from either dUMP misincorporation in place of dTMP during DNA synthesis or pre-existing cytosine deamination (56). Uracil-DNA glycosylase was first detected in E. co/i, and has since been isolated from a wide variety of biological sources through out the three kingdoms (58). Thus, uracil-DNA glycosylase is ubiquitously distributed in nature, and reflects the biological significance of this enzyme in the living organism. The uracil-DNA glycosylase family-i enzymes, named for their homology to the first detected uracil- DNA glycosylase in Escherichia coil (56), are highly conserved among different species, and are thought to have evolved from a common ancestor (59).

1.2.1. Escherichia coil Uracil-DNA Glycosylase

E. co/i uracil-DNA glycosylase (LJng), originally discovered by Lindahl and co-workers, was the first uracil-DNA glycosylase identified and purified to apparent homogeneity in vifro (56). The native enzyme is monomeric with an apparent molecular weight of 24,500 Daltons as determined by hydrodynamic properties and sodium dodecyl sulfate polyacrylamide gel electrophoresis (56). The gene (ung) encoding Ung was later cloned by Varshney and coworkers (60). The open reading frame of the ung gene was found to encode a protein of 229 amino acids with a deduced molecular weight of 25,664 Daltons (60). Detailed protein N-terminal sequence analysis of the native enzyme revealed that the N-terminal methionine of Ung was post-translationally removed (60). Overexpression of the ung gene and subsequent purification of large quantities of Ung have helped to elucidate the biochemical properties of this enzyme (61). In vitro, Ung has a broad pH optimum around 8. It does not require divalent cations or other cofactors for its activity and is active in the presence of EDTA (56). In addition, Ung efficiently releases uracil residues from single-stranded DNA as well as from double-stranded DNA containing U'A base pairs or UG niispairs, and has no associated endonuclease or AP lyase activity (56). Excision of uracil residue by Ung produces a free uracil base as well as an apurinic/apyrimidinic (AP) site (56). The former product (free uracil) was shown to

inhibit Ung activity with an apparentK1= 1.2 x 1oM (56). The latter product (AP site) was found to inhibit Ung activity at a concentration of approximately -4 p.M, which was -'100-fold lower than that of free uracil (57). TheKmvalue of Ung for dUMP residues in DNA and the turnover number of Ung were determined as approximately 4 xl 0 M and -800 uracil residues released per minute, respectively (56). Detailed characterization of substrate preference and specificity for Ung has revealed that Ung removes uracil residues more efficiently (-'2.5-fold) from single- stranded DNA than from double-stranded DNA (5 6,62). Moreover, Ung was found to preferentially remove UG mispairs over UA base pairs from double-stranded DNA (3 4,62). Although Ung preferentially excises uracil residues from single-stranded DNA substrates, excision of uracil from tetraloop hairpins was poor (63). This may be due to the unfavorable backbone and nucleotide conformation for efficient binding by the enzyme. Indeed, an NIvIR study on these hairpin DNA structures has shown that the unfavorable DNA phosphate backbone geometry resulted in a poor Km value, whereas the unfavorable nucleotide conformation resulted in a poorVmaxvalue of Ung (64). Addition of E. coli single-stranded DNA binding protein, which helped to melt the local hairpin structures, greatly improved Ung catalytic efficiency in removing uracil residues from these DNAs (63). The rigidity of the uracil-binding pocket of Ung restricts the number of substrates that are recognized. Ung does not release uracil from deoxyuridine, free dUMP, uridine, dUTP or RNA, nor are bromouracil, pyrimidine dimers, or deaminated purine residues removed by Ung (56). However, Ung is able to hydrolyze 5-fluorouracil, 5-hydroxyuracil, and 5,6-dihydroxyuracil in DNA (65-67). These results suggest that a larger modifying group at the C5 position of the uracil 10

ring, such as 5-methylcytosine or 5-bromouracil, are sterically hindered from entering the active site, and, therefore, are not subject to enzyme-catalyzed hydrolysis. In order to defme the substrate specificity of Ung, Varshney and van de Sande (60) constructed synthetic oligonucleotides of various lengths that contained a uracil residue(s) at site-specific positions. They demonstrated that Ung was unable to excise a uracil residue located at the 3'-terminal position of an oligomer or a 5'-terminal U from an oligomer, if the 5'-end was not phosphorylated (68). In contrast, uracil residues in the context pd(UN)p or pd(UNN) could be excised (68). Their fmdings indicated that not only was the oligonucleotide, 5' -pd(UN)p-3', the smallest uracil- DNA substrate recognized by Ung, but also that the 5 'phosphate and the two 3' phosphate groups adjacent to the uracil residue were critical for Ung-DNA interactions (68). Using synthetic oligonucleotides containing a pyrophosphate internucleotide bond near or adjacent to a uracil residue, Purmal and co-workers showed that the phosphodiester bond 3' immediately adjacent to a uracil residue was crucial for substrate binding by Ung (69). Similar results were observed using methylphosphonate substitutions in the phosphodiester linkage near or adjacent to a uracil residue (70). The rate of uracil removal by Ung from different positions in double-stranded DNA is also influenced by the sequence context surrounding the target uracil residue (71). For example, the rate of uracil excision from defined UA base pairs in different positions of double-stranded M13 DNA varies more than 15 fold, where uracil residues flanked by adenine or thymine at 5' and 3' sides are removed more efficiently than those by guanine or cytosine (71). Thus, the consensus sequences for efficient uracil removal by Ung are 5'-(AIT)UA(AIT)-3', while the sequences for poor uracil removal are 5'-(G/C)U(T/G/C)-3' (71). These results suggest that local DNA flexibility and DNA sequence context may influence the mutation frequency at specific sites (hotspots) in genomic DNA. Proteins that bind specific sites or sequences in DNA locate their targets by one of two general mechanisms. A processive search mechanism is established when the protein or enzyme remains associated with a particular DNA strand and locates 11

sequential target sites by facilitated diffusion (72). A distributive mechanism occurs when enzyme dissociates from the DNA following catalysis or site-specific binding and is subjected to random three-dimensional diffusion before locating the next target (72). Three studies have addressed the mechanism by which E. coli Ung locates uracil residues in DNA (62,73,74). Using a covalently-closed circular duplex DNA substrate (Form I pBR322) and treated simultaneously with Ung and T4 endonuclease V, Higley and Lloyd demonstrated that Ung acted with partial processivity at 50 mM NaC1 (73). In their experiments, processivity was determined by measuring the conversion of uracil-containing form I DNA to form III (linearized) DNA. Form III DNA was only produced following removal of closely spaced uracil residues on opposite DNA strands, followed by AP-site cleavage by T4 endonuclease V to produce two adjacent single-strand breaks. Significant accumulation of Form III DNA accompanied by unreacted Form I DNA was observed which was indicative of a processive mechanism. Adopting a different approach, Bennett and co-workers (62) constructed a concatemeric polynucleotide substrate that contained site-specific uracil residues at intervals of 25 nucleotides on one DNA strand. If Ung utilized a distributive mechanism, then the DNA fragments released by uracil excision (and subsequent AP-site cleavage) would tend to be large, since the probability that two adjacent uracil-residues were excised would be low. However, the majority of DNA fragments released were 25-mers, which indicated that Ung was able to excise successive uracils on one DNA strand (62). Similar results were obtained for both UA- and UG-DNA substrates. The addition of NaC1 ( 50 mM) reduced the amount of 25-mer detected, suggesting a transition from a processive to a distributive search mechanism. Overall, the results of Bennett et al. (62) were consistent with those reported by Higley and Lloyd (73). In contrast, Purmal and co-workers (74) utilized a linear double-stranded concatemeric polynucleotide with uracil residues spaced every 20 nucleotides along both DNA strands to study the Ung "search" mechanism, and concluded from their results that Ung used a distributive mode of action to locate uracil residues (74). However, the interpretation of their results was complicated by 12

the fact that the uracil residues were actually spaced only ten nucleotides apart relative to those on the opposite strand. Furthermore, the uracil target sites were located in a poor consensus sequence (5'-CUT-3') for efficient uracil removal, and the product analysis was performed after significant levels of uracil (>50 %) had been hydrolyzed from the DNA substrate (74). Under these circumstances, the concentration of AP- sites would be as high, or higher, then the concentration of uracil sites. Since Ung binds strongly to AP-sites, an assessment of processivity cannot be made for the experiments of Purmal et al.

1.2.2. Escherichia coli Double-stranded Uracil-DNA Glycosylase

The other enzyme with known uracil-DNA glycosylase activity in Escherichia coli is E. coli double-stranded uracil-DNA glycosylase (Dug). Using a database search with a core catalytic region of human thymine-DNA glycosylase (TDG) as a probe that is capable of processing UG but not T'G mispairs in DNA, Gallinari and Jiricny have identified TDG (See Section 1.3.1) homologs in gram-negative bacteria Escherichia coli and Serrattia marcescens (75). The enzyme sequence from E. coli open reading frame (ORF1-69), located between nucleotide positions 3,212,608 and 3,213,115 on theE. coli chromosome, is smaller than TDG by 120 and 100 residues at the amino and carboxyl terminus, respectively. However, the core regions display greater than 30% sequence identity (75). Recombinant E. coli enzyme expressed either in reticulocyte lysates or in E. coli was resistant to inhibition by Ugi and showed no activity on U'A base pairs or uracil in ssDNA (75). Since the E. coli enzyme was specific for uracil-containing duplex DNA, it was originally named double-stranded uracil-DNA glycosylase (dsUDG) (75). Also, because dsUDG preferentially excised UG mispairs, it was later named as mismatch specific uracil-DNA glycosylase (Mug) in the literature (76). Likewise, Saparbaev and Laval refer to dsUDG as ethenoC-DNA glycosylase (cCDG) because of its high excision activity against ethenocytosine in duplex DNA (77). However, in compliance with traditional E. coli nomenclature, 13

Sung and Mosbaugh have proposed that dsUDG (double-stranded llracil-DNA glycosylase) be referred to as Dug (78). Using a duplex 34-mer oligonucleotide containing 3,N4-ethenocytosine (EC),

uracil, or thymine to compare the substrate specificity of purified Dugin vitro, Saparbaev and Lava! showed that Dug acted on DNA substrates in the order of cCG > UG>> T'G (77). The catalytic efficiency(kcatfkm) of Dug acting on .CG (0.38 mm/nM) base pairs was 50 and 22,353-fold more efficient than that of UG (0.0077 mm/nM) and T'G (1.7 xl 0 mm/nM) mispairs, respectively (77). These results suggest that the excision of thymine residues in TG mispairs by Dug may not have a real biological significance because of its extremely low catalytic efficiency on this substrate. Accordingly, 3,N4-ethenocytosine, not uracil, may be a primary substrate for

E.coil Dug (77). This point of view was later supported by the studies of the repair of U'GorTGmispairs in Ung, Dug, or Vsr (very short repair)-proficient and -deficient isogenicE.coil cells. In this study, a genetic reversion assay involving a defective kanamycin resistance gene was used to assess the role of Dug in avoiding spontaneous C to T mutations. Since the reversion to kanamycin resistance in this assay results only from C to T mutations at a site for cytosine methylation, either a 5-methylcytosine to T change or C to U to T change can be studied with this system.UGorTGmispairs are the intermediates in these genetic reversion pathways, and hence any excision of T or U by Dug should reduce C to T mutations (79). The results showed that inactivation of dug had no effect on C to T or 5-methylcytosine to T mutation frequency inE. coil. In contrast, inactivation ofungor vsr significantly increased C to I or 5- methylcytosine to T mutations, respectively. Since Ung and Vsr, a DNA mismatch endonuclease, are known to repairUGorT'Gmispairs, respectively, inE. coil,these results suggest that Dug does not repairU'GandTGmispairsin vivo(79-8 1). In contrast to this report, Sung and Mosbaugh have provided evidence for the involvement of Dug in uracil-DNA repairin vitroandin vivo(78,82). An Ml3mp2opl4 DNA (form I) containing a site-specificUGmispair was used to evaluate whether Dug could initiate a uracil-mediated base excision repair in E. coil 14

NR8052(ung dug)cell extracts. The results revealed a time-dependent appearance of repaired form I DNA in the Ung-deficient cell, and addition of exogenous purified Dug to the cell extract stimulated the rate of repair. Additionally, the rate and extent of complete uracil-DNA repair was measured using an Ml 3mp2 lacZa DNA-based reversion assay together with Ung or Dug-proficient and -deficient isogenicE. co/i cell free extracts (82). In reactions utilizingE. co/iNR8052(ung dug)cell free extracts, approximately 20% of the uracil-DNA was repaired, and these repair reactions were insensitive to inhibition by the PBS2 uracil-DNA glycosylase inhibitor protein. In the reactions withE. co/iNR805 1(ung dug)cell extracts, -P80% repair was observed, and the rate of repair in these Ung-proficient reactions was 5-fold greater than that in the Ung-deficient but Dug-proficient reactions. In contrast, uracil- DNA repair was not detected in reactions deficient in both Ung and Dug cells (E. co/i

BH1 58). Overall, these results suggest that Dug involves UG mismatch repairin vivo as well asin vitro. The crystal structures of free Dug and Dug in complex with duplex DNA containing a non-hydrolyzable deoxyuridine analogue, 1 -(2 'deoxy-2 'fluoro-3-D- arabinofuranosyl)-uracil, mismatched with guanine have revealed a structural and functional similarity to Ung despite its lack of sequence homology with Ung (76,83). However, unlike Ung, which excludes bases other than uracil from its active site, the topologically relevant amino acid residues in the base-recognition pocket of Dug can accommodate larger modifications to a uracil base (83). This non-specific pyrimidine- binding pocket provides a rational basis for the observation that Dug excises exocyclic DNA adducts, such as N4-ethenocytosine (SC), 1 ,N2-ethenoguanine (&G) (84), and 8-

(hydroxymethyl)-3,N4-ethenocytosine (85) from duplex DNA. Analysis of thecrystal structure of Dug-DNA complex shows that interactions of Gly 143, Leu- 144, and Arg 146 with widowed guanine on the opposite strand opposite to uracil analogue may provide significant enzyme specificity for UG and TG mispairs but not U'A or TA base pairs or ssDNA (76). Consistent with this hypothesis, binding of Dug to an abasic 15

site opposite a widowed guanine is 10 times stronger than to an abasic site opposite a widowed adenine (76).

1.2.3. Mammalian Uracil-DNA Glycosylases

1.2.3.1 Nuclear and Mitochondrial Uracil-DNA Glycosylase

Mammalian mitochondrial and nuclear uracil-DNA glycosylase activities have been isolated and characterized from several biological sources including rat liver (86,87), calf thymus (88,89), and human cells (90-92). Biochemical characterization of homogeneous preparations of mammalian nuclear and mitochondrial uracil-DNA glycosylase revealed that the nuclear and mitochondrial forms of uracil-DNA glycosylase have apparent differences in size. For instance, the nuclear and mitochondrial species of rat liver uracil-DNA glycosylase have molecular weights of -'35 and -24 kDa, respectively (86). While the nuclear and mitochondrial species from human placenta have molecular weights of -'50 and --'18 kDa, respectively (91). These apparent differences are similar to some other reports that suggest a high molecular weight (--'30 to -'37 kDa) enzyme for the nuclear enzyme and a lower molecular weight (--'18 to --'20 kDa) for the mitochondrial enzyme (90,91,93,94). Nonetheless, biochemical properties of isolated mammalian uracil-DNA glycosylase from different resources have a high degree similarity with E. coil uracil-DNA glycosylase (Ung). First, the nuclear and mitochondrial uracil-DNA glycosylase from calf thymus have been shown to be monomeric and do not require any cofactor for activity (88). Secondly, these enzymes preferentially (-'2 fold) excise uracil residues from a single- stranded DNA over a double-stranded DNA (95). Thirdly, like E. coli Ung, mammalian uracil-DNA glycosylases are inhibited to varying degrees by the glycosylase reaction products, free uracil and abasic site-DNA. Uracil has been shown to inhibit mammalian nuclear or mitochondrial uracil-DNA glycosylase in a noncompetitive maimer at millimolar (mM) concentrations (87,90,93), whereas abasic 16

site-DNA competitively inhibited at micromolar (pM) levels (87,89). Fourth, the substrate specificity for bovine uracil-DNA glycosylase was found to be similar to that of E. coil Ung, since a -10-fold difference in excision rates was reported for good and poor consensus sequences for uracil removal, 5'-A/TUAA/T-3' and 5'-G/CUG/C-3', respectively (95). Finally, the expression of the human enzyme as a LacZa-humUNG fusion protein in E. coil complements E. coil ung mutants (96). These results suggest that the structure and function of uracil-DNA glycosylase in living organisms is evolutionarily conserved for efficient removal of uracil residues from DNA.

1.2.3.2. Human Nuclearand Mitochondriai Form Uracil-DNA Glycosylases and TheirCommonCatalyticDomain

The presence of uracil-DNA glycosylase activity in both nuclear and mitochondrial fractions of human cells in culture was originally reported in 1980 by Anderson and Friedberg (97). Subsequently, human nuclear and mitochondrial uracil- DNA glycosylase were partially purified from nuclear and mitochondrial extracts of different sources of human cells (90-92,94,98). However, there had been some controversy as to the number, identity, and subcellular localization of the enzymes in these early studies. The human uracil-DNA glycosylase isolated from blast cells of patients with acute myelocytic leukemia existed in a single form with a molecular weight around 30 kDa, and its activity was inhibited by the addition (10-25 mM) of

MgCl2,MnC12, CaCl2,NaCI, KCI, EDTA or EGTA (90). In contrast, the enzyme isolated from extracts of human placenta existed as a basic 29 kDa protein with a turnover number of-' 600 uracil residues/mm and its activity was stimulated -40-fold by 60-70 mM monovalent salt (92). The human uracil-DNA glycosylase in HeLa S3 cells was originally found as a single form, and the partially purified enzyme preferentially (-3 fold) excised uracil residues from single-stranded DNA relative to the corresponding double-stranded DNA substrate. The HeLa enzyme was inhibited by free uracil and 6-aminouracil, as well as by 5-azauracil (98). However, a subsequent report by the same group showed that this single enzyme existed in cellular 17 nuclei (70%), mitochondria (15%) and cytosol (15%), and when isolated from the extracts of the cells, it appeared as a major 50 kDa form and as a minor 18 kDa form (9 1). The two forms were both stimulated by 40-60 mlvi monovalent salt (91). The differences among these early studies may be due to the various degrees of proteolysis of the enzymes during protein preparations. However, despite these differences, the relatedness of the mitochondrial and nuclear enzymes was later demonstrated by antibody cross-reactivity, which suggested that the two enzymes shared similar antigenic determinants (99-10 1). The cloning of humanUNGgene by Slupphaug and co-workers (99) confirmed that nuclear (UNG2) or mitochondrial (UNG1) forms of human uracil- DNA glycosylase, are encoded by the same gene. TheUNGgene comprises 6 exons and 5 introns and was assigned to chromosome 12q23-q24.1 as confirmed by radiation hybrid mapping technique (102). In the human genome,UNGgene spans13.5Kb including the promoter, which contains a 5'-CpG island of 1.2 kb and is a very GC- rich but has a TATA-less promoter (102). Using aUNGpromoter-luciferase construct (pGL2-ProB), Haugh et al. (102) showed that methylation ofUNGpromoter at 5' CpG sites strongly reduced transcription. However, methylation of theUNGpromoter may not be important for the regulation of UNG expression in vivo. Results of Southern blot analysis onUNGpromoter methylation status from eighteen different cell lines with various levels of uracil-DNA glycosylase activity showed no significant difference in promoter methylation (102). These results suggest that the binding sites for the putative transcription factors in the UNG promoter region are maintained in an unmethylated state, regardless of the expression level (102). Transcription factors can have a positive or negative influence onUNGgene expression. The transcription factor E2F-1/DP-1-Rb complex was known to be a strong negative regulator ofUNG expression, while Spi was shown to act as a positive regulator (103). The mRNAs for UNG1 and UNG2 increased 2.5- and 5-fold, respectively, in late Gi/early S phase, and were accompanied by 4- to 5-fold increase in enzyme activity in synchronized HaCaT cells (103). The expression level of UNG1 and UNG2 mRNAs varies among 18

different tissues in the human body. UNG1 mRNA was found to have the highest levels in skeletal muscle, heart and testis, whereas the highest UNG2 mRNA levels were found in organs contain proliferating cells, such as placenta, colon, small intestine and thymus, as well as testis (103). UNG1 and UNG2 are generated from the alternative splicing and transcription

from different positions(PAandPB)in the first exon of UNG gene (104). The UNG1 sequence starts at codon 1 in exon lB and composes 304 amino acids, the first 35 of which are unique to this form, while UNG2 sequence starts at codon 1 in exon 1A and constitutes 313 amino acids, the first 44 of which are unique to UNG2 (104). The remaining 269 amino acids of UNG2 and T.JNG1 are identical and constitute the common catalytic domain of the enzyme (104). The unique N-terminal sequences in UNG1 and UNG2 are required for mitochondrial and nuclear localization, respectively, but not for catalytic activity (104). The unique N-terminal residues of UNG1 constitute a strong and complete mitochondrial localization signal (MLS), because residues 11-28 in UNGihave the potential of forming an amphiphilic helix, which is a typical feature for MLS (105). When the MLS of UNG1 was placed at the N-terminus of UNG2, the nuclear localization signal of UNG2 was overridden and the nuclear protein was directed to the mitochondria (105). Once UNG1 was imported to mitochondria, it was further processed by a mitochondrial processing peptidase and an unknown mitochondrial peptidase to form UNG1A29 (31 kDa) and UNGA75/77 (26 kDa), respectively (105). UNG1A29 was not inhibited by AP-sites; however, the TJNGA75/77 form was inhibited by AP-sites at micromolar concentration (105). The unique N-terminal sequence of UNG2 was reported to have several functions: 1) It contained a nuclear localization signal for nuclear sorting (104,105) and, 2) a conserved PCNA and RPA binding motif for binding PCNA and RPA (106). 3) It was required for stimulation of UNG2 activity byMg2(107), and 4) it contained serine/theonine phosphorylation sites (Thr3 1 and Ser4O), which might be important for enzyme activity and protein-protein interactions (107,108). 5) It might also interact 19

directly with human AP-endonuclease 1 (APE 1) (107); however, no data has been published on this subject to date. The common catalytic domain of UNG1 and UNG2 (residues 85-304 according to UNG1 numbering), designated as UNG, was originally discovered when Slupphaug and coworkers investigated the possible significance of the 77 amino acid presequence ofUNG1 for enzymatic activity (100). In the study, to obtain a better expression in E. coli and to examine how much of the N-terminal portion of UNG1 could be removed before its activity was lost, plasmid vectors containing a series of 5'- deletions (corresponding to N-terminal 28 to 107 amino acids of UNG1) were created using a standard exonuclease digestion technique. The clone harboring pUNGA84 encoded UNG* was found to have the highest expression inE.co/i as well as the highest relative enzyme activity after in vitro translation (100). Subsequently, the pUNGM4 was cloned into an expression vector (pTUNGA84), and homogeneous

TJNG* was then overproduced and purified fromE.co/i cell extracts. As analyzed by SDS-PAGE, the recombinant UNG* migrated as 27 kDa protein and had a specific activity of 8500 units/mg in the assay conditions containing 10 mM NaC1 (100). Increasing the salt concentration beyond 10 mM resulted in a gradual decrease in UNG* activity, and enzyme activity was essentially abolished at concentrations above 200 mM (100). The pH optimum for UNG* activity was between 7.7 and 8.0 and the pI of UNG* was found to be between 10.4 and 10.8 (100). In addition, the influence of DNA sequence context on uracil removal by UNG* was similar to that for human UNG isolated from human placenta (95). UNG* was demonstrated to remove uracil residues in single-stranded DNA or duplex DNA containing UG mispairs or U'A base pairs in the order: ssU >U'G >UA (100). In addition to uracil residues, TJNG* also excised such uracil analogues as 5-hydroxyuracil, isodialuric acid, and alloxan at relatively low rates compared to uracil (109). 20

1.2.4. Crystal StructuresofUracil-DNA Glycosylases

Amino acid sequence alignment analysis of uracil-DNA glycosylases (LTDGs)

from herpes simplex virus type-i (HSV-l),E. co/i, and homo sapiensrevealed that the E. coil enzyme is 56% identical to the human enzyme and 41% identical to the viral enzyme (110). Although the primary amino acid sequences vary to some degree among these three enzymes, the overall crystal structures of these enzymes are highly conserved as revealed by the superposition of the three enzyme structures (110-113). All known crystal structures of UDGs (110-i 12) are typically a single domain a/f3 folded proteins consisting of a central parallel four-strand t3 sheet with a f32-31-P3-134 topology, which is associated with eight a helices flanking at either side. The crystal structure of UDG from HSV-1 (110) was the first crystal structure of a uracil-DNA glycosylase family-i enzyme. The 1.75Acrystal structure of the HSV-1 UDG (110) (residues Leul7 to the C-terminus at residue Va1244) displays a single protein domain formed by a left-handed coil of four helices leading into an alternating 13-a-3 structure. The twisted four parallel J3 sheets (2-1-3-4 topology) lie at the heart of the structure, and the N- and C-termini of the protein lie on opposite sides of the central 3 sheet. The charge distribution on the surface of the enzyme is highly anisotropic and shows a slight excess of basic residues on one of the large faces of the enzyme. A winding channel runs across this positive-charged face with a distinct pocket near one end of the channel which is formed by the residues Pro86, G1n87, Asp88, Tyr9O, and Phe101 (110). All of the residues involved in the formation of the pocket are located at the C-terminal ends of the first and second-strands and are conserved in E. coli and human enzymes (110). Like the viral enzyme, the structure of

E.coli Ung solved by Putnam and co-workers (114) is an a/J3 fold protein that contains a parallel four-stranded f3-sheet in the center flanked on both sides by eight a- helices. The N- and C-termini lie on opposite sides of the central J3 sheet. The center doubly wound 2-i-J33-4 sheet divides the single-domain enzyme into two distinct halves. The first half contains al-5 and p1-2 (residues 5-140) and the other half 21

consists of a6-8 and 133-4 (residues 141-227). A -45A wide groove is formed between the 131 and 133 sheet. This groove is lined with positively charged and functionally conserved amino acid residues, and is similar to the viral enzyme. The structure of the common catalytic domain of human uracil-DNA glycosylase (UNG*) (111) resembles that of HSV-landE.coil. The enzyme consists of four central f3 strands (132-131-133-134 topology) surrounded on either side by a total of eight a-helices. The N- and C-terniini

lie on opposite sides of the central13sheet, and there are no disulfide linkages found in

the structure. The DNA binding groove is formed at one end of the central13sheet and has a -2 1Adiameter at the rim. It is surrounded by the functionally conserved basic residues His148, Arg2lO, Lys218, Arg220, Lys251, and Arg276 (111).

Although the overall structures of three UDGs from HSV- 1,E.coil, and human are similar, significant structural differences are observed (112). Because of the insertion of the non-conserved G1y34 in the 131 strand, the al and a2 helices ofE. coil Ung are shifted one- half and one-quarter turn, respectively, relative to the HSV- 1 and human enzymes. The insertion of 1le175 inE. coilUng and 1le256 in human UNG in the a helix preceding the 134-strand of both proteins results in Ca shifts of up to 2.2A in this region relative to HSV-1 enzyme. Additionally, the insertion of G1y199 inE. coil Ung and Gly277 in human UNG in the loop preceding a8 helix of both proteins results in a backbone shift as large as 6.4Acompared to the viral enzyme. Finally, the last six to seven residues (residues 223-229) at the C-terminus of E. co/i Ung go beyond the C-termini in the viral and human enzymes and therefore extend as a unique tail in the structure. Co-crystal structures of UNG* in a complex of duplex DNA containing either U'A, 4'-S-dUA and 2'-dijrU.A base pairs, or U'G mispairs reveal that uracil- containing DNA lies across a positively charged groove at the C-terminal end of the central four-stranded 13-sheet of the enzyme with enzyme-induced DNA distortions localized at the flipped-out uracil nucleotide (115-117). The Ni-Cl' glycosylic bond linking the uracil base and deoxyribose has been cleaved in these structures (115,116), and the helical space of the flipped-out nucleotide in the DNA is replaced by the 22

intercalation of the leucine-loop between 134 and a8 into the DNA minor groove. The uracil base binds within the uracil specificity pocket, stacks with Phe 158, and forms hydrogen bonds via its 04, N3 and 02 atoms to five different residues in the pocket: the amide N group (-NH) of Phe 158; the side chain of Asn204; the backbone atoms of G1n144 and Asn145, and the imidazole of His 268, respectively. Also, Tyr147 packs against the uracil CS position and, as a consequence, selects against thymine, which contains a methylgroupat this position. The deoxyribose binds within the uracil specificity pocket and forms hydrogen bonds from 04' and 01' to His148 and Asn145, respectively. Analysis of the distorted DNA structures shows that the distance between the phosphates flanking the flipped-out uracil nucleotide is compressed by 4A(from

-42 to 8A).This compressed phosphate backbone results in bending of the DNA by -

450, and the kinking of the DNA helical axis by 2A.By contrast, the bound DNA both 5' and 3' to the flipped-out nucleotide retains a B-form DNA structure, with an average of 10.7 base pairs per turn. Detailed analysis of the UNG*DNA co-crystal structure indicated the total UNG*DNA interface was relative small, and over half (- 63%) of the interactions involved the extrahelical uracil, deoxyribose and 5'-phosphate (115,116). Notably, the enzyme-DNA interface lacked direct and water-mediated DNA phosphate contacts with basic amino acid side chains. Of five arginines and eight lysines localized to the positively charged active site surface, only Arg276 directly contacted the DNA. Interestingly, this direct interaction between Arg276 and DNA was not noted in the co-crystal structure containing a U'G DNA mispair (115). As for the enzyme-DNA interactions, the uracil 5'-phosphategroupbound near an a-helix dipole and formed hydrogen bonds with the NH and OHgroups ofSer169, and the uracil 3'-phosphate groupformed hydrogen bonds with the OHgroupof Ser270 and Ser273. One nucleotide further downstream (3') of the uracil residue, the second phosphategroup, which was also stabilized by an a-helix dipole, interacted with the Ser247-amide hydrogen and -OHgroupsas well as with the His268-aniide hydrogen. In addition, N3 of the second base (guanine) downstream (3') of uracil directly interacted with iN of 23

Arg276 through hydrogen bonding. Inspection of the enzyme-DNA interface showed that the leucine intercalation ioop penetrated the DNA minor groove, and the side chain of Leu272 extended past the abasic site and into the DNA major groove. Structure-based sequence alignments of uracil-DNA glycosylases from a variety of biological sources have identified five highly conserved motifs key to enzyme activity. In the UNG1 amino acid sequence, these motifs are: 1) the catalytic water activating ioop (145-DPYH-148); 2) the proline-rich loop (165-PPPPS-169); 3) the uracil specificity p2-region (201-LLLN-204); 4) the Gly-Ser loop (246-GS-247), and 5) the leucine intercalation loop (268-HPSPLSVYR-276). Comparison of the crystal structures of free IJNG* and UNG*in complex with uracil-DNA revealed that UNG* underwent a global conformational change froman "open" unbound state to a "closed" DNA-bound state upon binding uracil-DNA (115-117). Key conformational changes occurred in the loops that interact with DNA, particularly the leucine-loop, which shifted towards the enzyme active site. The concerted loop movements centered on a f3-zipper, which consists off31, 133, and a part of thecloop (the end of 131 to beginning of the a4). Closure of the 13-zipper effectively clamped the enzyme around the flipped-out uracil nucleotide and was accompanied by insertion of leucine-loop into the DNA minor groove. These discrete loop movements brought key amino acid residues into functional positions where they could interact with the flipped-out uracil nucleotide and create the catalytically competent enzyme active center. A similar13- zipper mechanism was proposed for E. coli Ung (112,114).

1.2.5. Mechanism ofAction

The enzyme-DNA co-crystal structures and mutational studies of E. coli Ung and human UNG* suggested that the general mechanism used by family-i uracil-DNA glycosylases (UDGs) to excise uracil residues from DNA involved two distinct events: 1) initial damage detection and uracil nucleotide flipping, and 2) catalytic mechanism ofNl-Cl'-glycosylic bond cleavage (112,115-119). 24

Based on several UNG*DNA co-crystal structures, Slupphaugetal.(115) and

Parilthet al.(116) proposed that UNG bound DNA via electrostatic interactions between the positively charged active site face of the enzyme and the negatively charged DNA phosphate backbone, which facilitated correct orientation of the enzyme active site along the DNA. Minor groove scanning coupled with DNA phosphate backbone compression led to initial uracil detection followed closely by uracil nucleotide flipping. Slupphaug and co-workers (115) hypothesized that UNG* excised uracil residues from DNA via a "Pinch-Push-Pull" catalytic mechanism. The proposed mechanism consisted of three steps: 1) UNG* scans the DNA minor groove for uracil using the minor groove reading head, which is composed of Tyr275 and Arg276 of the leucine intercalation ioop. In binding DNA, UNG* compresses (pinch) the duplex DNA phosphate backbone via the 4-Pro-Ser and Gly-Ser loop. 2) Upon locating a uracil residue, the side-chain of Leu272 is inserted into the DNA minor groove, pushing uracil nucleotide out of the DNA helix from the major groove and occupying its helical space. 3) The partially flipped-out uracil nucleotide is caught and pulled deeper into the conserved uracil specificity pocket, where the uracil base is recognized and the glycosylic bond is cleaved. The "Pinch-Push-Pull" catalytic mechanism was first supported in a study on the kinetic mechanism of E. coil Ung by Stivers (118). In this study, duplex DNA substrates were constructed that contained a non-hydrolyzable uracil analogue, 2'-a and 2'-13 fluorine isomers of T-fluoro-2'-deoxyuridine, positioned adjacent to a fluorescent nucleotide reporter group, 2-aminopurine (2AP) to measure the equilibrium DNA binding and nucleotide flipping in the absence of glycosidic bond cleavage. When E. coil Ung bound these DNA substrates, a 4-8-fold 2AP fluorescence enhancement, which resulted from enzymatic flipping of the adjacent 2'-a- and 2'-J3- fluoro-2'-deoxyuridine, was observed. In addition, the authors monitored the change in intrinsic enzyme tryptophan fluorescence whenE.coii Ung bound uracil-DNA. The results showed that intrinsic enzyme fluorescence decreased by 1.8-fold whenE. coil Ung bound AUAfFGT DNA. However, enzyme fluorescence remained unchanged 25 when Ung bound normal (non-uracil-containing) duplex DNA under saturating conditions in equilibrium DNA binding measurements. Hence, Stivers and co-workers

(118) reasoned that uracil-specific binding altered the conformation ofE.coli Ung, which resulted in quenching of intrinsic tryptophan fluorescence. Based on the change in conformation observed for free UNG* and UNG* bound to uracil containing-DNA

(113), and given the structural similarity between human UNG* andE.co/i Ung

(114), Stivers and co-workers proposed thatE.co/i Ung also underwent a conformational change upon binding uracil-containing DNA, and that this change affected the fluorescence properties of the enzyme's tryptophan residue(s) by changing the solvent environment (118). While the identity of the tryptophan residue(s) responsible for this fluorescence change has not been determined, of the six tryptophan residues inE.co/i Ung, Stivers et al. (118) suggested that W141 and W164 might be responsible for the fluorescence change, since they were situated close to the enzyme active site. Further kinetic studies suggested that site-specific DNA binding occurred by a two-step mechanism. The first step involved formation of a weak nonspecific enzyme-

DNA complex (ES) with aKd 1.5 .iM. The second step involved formation of a specific enzyme-DNA complex (E'F) that resulted in kinking and a rapid and reversible flipping of deoxyuridine into an extrahelical state with 1200 s1. This later step was followed closely by a change in enzyme conformation that was associated with insertion of the leucine-intercalation loopintothe DNA minor groove and could be monitored by quenching of the enzyme's intrinsic tryptophan fluorescence. Thus, uracil flipping occurred concurrently with enzyme conformational change and insertion of Leul 91 side chain into the DNA minor groove. Using the same experimental approach, Jiang and Stivers (120) refined the mechanism for Ung DNA binding and uracil flipping to include three steps: 1) E + 54-+ES, the nonspecific 4 DNA binding and dissociation step, was characterized byk1= 220 .iM' andk1 =

600s12) ESE-*EF, reversible uracil flipping into an extrahelical state, with kj = 700 sand kfl = 180 s4); and 3) EF4+E*F, conformational docking ofE.co/i Ung around the flipped-out base, with = 350 s_i andkcoç=100 s_i. The insertion of Leu 191 side chain into the DNA minor groove occurred after uracil flipping into an

s_i extrahelical state(kcoflf,= 350 base pair, Wong and co-workers (121) proposed an alternative "pinch-pull-push" catalytic mechanism for uracil-excision from the DNA. Stopped-flow fluorometry was used to monitor 2AP fluorescence and enzyme intrinsic fluorescence quenching, which define the kinetics of uracil flipping and enzyme confonnational change/leucine-loop insertion, respectively. Their results showed that 2AP fluorescence enhancement preceded enzyme intrinsic fluorescence quenching because the rate of flipping, kii = 1100s_i,was faster than the rate of conformational change, = 300 Therefore, they concluded that the insertion of the conserved leucine-loop into the base stack occurredafterthe uracil nucleotide was pulled into the enzyme active site pocket. In addition, using a rapid chemical-quenched-flow method to investigate the internal dynamics of the steps leading up to glycosylic bond cleavage, they determined that for each uracil flipping event, there was approximately a 33 % probability that the uracil nucleotide would be pulled into the enzyme active site pocket. After docking in the active site, positive interactions between the uracil base and conserved amino acid side chains would trigger a conformational change. Thus, Wonget al.(121) concluded that uracil flipping and enzyme conformational change occurred prior to glycosylic bond cleavage. They proposed that insertion of the leucine-loop into the minor groove acted more as a doorstop to prevent the return of the flipped-out uracil nucleotide to the base stack, than as a lever to pry the uracil nucleotide into an extrahelical position. 27

By increasing the enzyme pocket/base stack uracil partition coefficient, leucine-loop insertion facilitated productive capture of the uracil in the enzyme active site.

Mutational analysis on the conserved residues of human UNG* andE. coli Ung (111,120,122) implicated in the "pinch-pull-push" catalytic mechanism further substantiated the role of these amino acids in the enzyme catalytic mechanism. Mutations of serine residues in the Pro-Ser and Gly-Ser loops of UNG* (S169A and

S270A, respectively) that are proposed topinchthe DNA phosphodiester backbone greatly diminished both the DNA binding by up to 25% and enzyme catalytic activity by 95% (111). Similarly, mutation of the conserved serine residues in the Pro-Ser and

Gly-Ser loops ofE. co/iUng, S88A and S189A, resulted in 9.2- and 4.8-fold reduction in DNA binding, respectively, while the S88A:S189A double mutant exhibited a 65- fold reduction (120). Mutation of the conserved Leu272 residue of the leucine intercalation loop of UNG*, proposed to push the uracil nucleotide out of DNA major groove, to alanine (L272A) or arginine (L272R) resulted in the reduction ofkcatby 80% for both single- and double-stranded uracil-containing DNA (115). Similar mutational effects were observed for the corresponding amino acid in E. co/i Ung, Leul9l. The mutation L191A and L191G showed a 7.3- and 39-fold, respectively, reduction in DNA binding affinity for the duplex 2'-3-fluoro-2'-deoxyuridine/2- aminopurine-containing DNA (120). In particular, nucleotide flipping, evaluated by the 2AP fluorescence enhancement when the adjacent uracil moiety is extrahelical, was significantly reduced in both L191A and L191G mutants (120). In addition, Ung intrinsic fluorescence quenching was totally abolished in the L191A and L191G mutants (120). These results suggested that insertion of leucine intercalation ioop into the DNA minor groove in both L191A and L191G mutants was compromised, and that the two mutants did not proceed to the global conformational change that results in tryptophan fluorescence change. Therefore, these two L191 mutants were unable either to push the uracil nucleotide out of the DNA major groove or to prevent the return of the flipped-out uracil nucleotide, resulting in a the reduction of nucleotide flipping and DNA binding (120). The defects in DNA binding and nucleotide flipping caused by Leul9l mutations could be rescued by putting a pyrene base opposite the uracil in the duplex DNA substrate (122). The pyrene base was shown to act as a molecular wedge to push the deoxyuridine from the base stack, thereby pre-organizing the deoxyuridine nucleotide in an extrahelical state (122). Presumably, the pyrene base mimics the proposed function of Leui9l and compensates for the Leui9l mutations (122). Finally, mutations (N204Q, N204V, and H268L) at Asn204 and His268 residues in the uracil specificity pocket of UNG*, that form hydrogen bonds with flipped-out uracil and are proposed to pull deoxyuridine out of DNA stack, perturb Asn204 and His268 hydrogen bond formation with the uracil base and thus reduce DNA binding and enzyme catalytic activity (111). Similar mutational effects are also found in the corresponding amino acid mutations in E. coil Ung. The N123G mutation was shown to bind 2'--fluoro-2'-deoxyuridine/2-aminopurine-containing DNA 16-fold more weakly than the wild-type Ung, while Hi 87G was found to have a 2-fold reduction in the DNA binding affmity for the same DNA substrate (120). Taken together, these results demonstrate the importance of the key residues in the conserved motifs of UNG* and E. coil Ung that are involved in the "pinch-pull-push" catalytic mechanism. The uracil specificity pocket forms hydrogen bonds and base stacking interactions with the uracil base through the sequence conserved residues, Asn204, His268 and Phe 158, and orients the Ni-Cl' glycosylic bond for hydrolysis (115,116). In addition, the side chain hydroxyl group of Tyrl 47, located just outside the active site pocket sterically excludes other pyrimidine bases and prevents non-uracil base cleavage (110-112,115,116). Substitutions of Tyr147 with Ala, Cys, or Ser result in enzymes that have cytosine-DNA glycosylase (CDG) activity or thymine-DNA glycosylase (TDG) activity, respectively (123). Based on the biochemical and structural studies of herpes simplex virus type-i (HSV-1), E. coil, and human uracil- DNA glycosylase, two reaction mechanisms have been proposed for the cleavage of the N-glycosylic bond linking uracil to deoxyribose in DNA. The first mechanism 29

involves sequence-conserved histidine and asparagine residues and the addition of an activated water molecule to the anomeric (Cl') carbon to sever the N-glycosylic bond. In E. coli Ung, the Ni-Cl' glycosylic bond is polarized by electron density withdrawal from the deoxyribose Cl' through Ni into the uracil ring. The developing negative charge is distributed about the uracil ring by resonance effects involving oxyanion formation at 02 and 04 and stabilized by hydrogen bonding with the side chains of His 187 and Asni 23, and the main-chain Gln-63 amide. Asp64 of the water-activating loop formed by 64-Asp-Pro-Tyr-His-67 serves as a general base and activates a water molecule associated with the active site, which subsequently attacks the Cl' position of deoxyribose (Figure 1) (119,124). The second mechanism entails step-wise dissociation (SN1-lilce), which involves complete breakage of the N-glycosylic bond and generation of a discrete oxacarbenium ion-uracilate anion intermediate, followed by the attack of the nucleophilic water at Cl' of the oxocarbenium ion (Figure 2). Werner and co-workers (125) performed kinetic isotope effect studies using E. coli Ung and variously labeled (light vs. heavy) uracil/deoxyribose nucleotides situated in a 4-mer oligonucleotide. The kinetic isotope effects were calculated from the observed rates of uracil release and used to interpret the transition-state structure of the uracil-deoxyribose reactants. The results showed that: 1) The N-glycosylic bond linking uracil to deoxyribose was completely broken during the transition state; and 2) the anomeric carbon contained significantsp2character, and the deoxyribose sugar adopted a mild 3'-exo sugar pucker in the transition state (125). This stepwise dissociative mechanism was further supported by the studies using oxacarbenium ion transition state mimics, such as 1- aza-2'-deoxyribose, to study the mechanism of catalysis byE. coli Ung (126,127) 30

Figure 1. Scheme of the reaction mechanism for cleavage of the Ni-Cl' glycosylic bond byE. coil Ung. Resonance forms of the proposed hydrolysis intermediates in the uracil-excision are illustrated. The Ni-Cl' glycosylic bond is polarized by electron density withdrawal from the deoxyribose Cl' through Ni into the uracil ring. The developing negative charge is distributed about the uracil ring by resonance effects involving oxyanion formation at 02 and 04 and stabilized by hydrogen bonding with the side chains of His- 187 and Asn- 123, and the main-chain Gln-63 amide. The catalytic water is activated by Asp-64 in the water-activating loop formed by 64-Asp- Pro-Tyr-His-67 (116,124). 31

1 H / p / fr

i

Y k

H

J 32

Figure 2. Interactions leading to stabilization of the oxocarbenium ion-uracil anion intermediate in the E. coli Ung active site. Stabilization of the transition intermediate is thought to arise from the enforcement of stabilizing hyperconjugative effects in the reactant and transition states using the serine phosphodiester interactions, and electrostatic stabilization of the oxocarbenium ion by the anionic uracil leaving group and Asp64 (125). 33

I.] iii1TM

Figure 2 34

A hybrid quantum-mechanical/molecular-mechanical (QM/MM) model of UNG* catalysis also supports this stepwise mechanism (128). An SN1-like reaction mechanism for the cleavage of the N-glycosylic bond linking uracil to deoxyribose in DNA was deduced from co-crystal structure of UNG* in complex with duplex xU-DNA (117). In this enzyme-DNA complex, the enzyme active site residues Asn204 and G1n144 form hydrogen bonds with llrU 04 and 06, respectively, and rotate the uracil ring900 on its Ni -C4 axis to a position almost halfway between anti andsyn conformation, thereby flattening the pucker of the uridine deoxyribose to a mild C3'-exo, and raising the glycosylic bond to a semiaxial position (Figure 3A). According to Parikh et al. (117), the deformation of deoxyribose enables ps*overlap (41) with 04t, which is termed the anomeric effect. They hypothesize that a second stereoelectronic effect is achieved via the glycosylic bond rotation and pyraniidalization ofNl (the glycosylic bond is bent approximately 50° out of the plane of the uracil ring), which together create electron orbital overlap between the glycosylic bond and the 02 and 04 carbonylitsystems of the uracil base. Thus, the enzyme active site center couples the two stereoelectronic effects by distorting the uracil nucleotide structure in order to increase sequential linear orbital overlap from the deoxyribose 04' to the uracil 02. Efficient electron transposition through orbital overlap makes the uracil anion a good leaving group and promotes glycosylic bond cleavage (Figure 3B). Parikh et al. (117) proposed that glycosylic bond cleavage most likely proceeded via a dissociative, SN1 -like process in which the leaving group, the enolate anion of uracil, was protonated at Ni subsequent to bond cleavage to give the more stable amide. 35

Figure 3. A. Deoxyuridine (gray carbon tubes, red oxygens, blue nitrogens) in DNA is severely distorted by the UNG* active center to achieve the observed conformation of the iU (orange carbon tubes, red oxygens, blue nitrogens). The left side of the large arrow is deoxyuridine in the conformation normally found in DNA. The arrow implies the observed flipping of the substrate nucleotide out of the DNA helix, which results in the altered position of the 5P. When flipped into the TJNG* active center (right side of large arrow), the uracil ring is rotated 9O° on its N1-C4 axis to an angle of 177°. Furthermore, the deoxyribose sugar of the enzyme bound substrate is flattened to a mild C3'-exo, which raises the uracil to a semiaxial position. The normally trigonal planar 1-position of uracil is strained to an almost tetrahedral geometry. The small arrows indicate the steric hindrance, which causes the deformation at the uracil 1- position. B. Structure-based reaction mechanism for Ni-Cl' glycosylic bond cleavage. 1. A simplified valence-bond representation of the glycosylic bond dissociation. The three electron pairs are transposed and are involved in orthogonal orbitals. 2. In the normal anti-conformation of deoxyuridine, the a*orbital involved in the anomeric effect and the it-orbital of the C2=O bond are orthogonal to one another, thus preventing orbital overlap. 3. Severe distortions of the deoxyribose and the glycosylic bond in the strained conformation of deoxyuridine enforced by the UNG* active center align the pairs of atomic orbitals participating in each electron transposition, thereby electronically coupling the anomeric and a-itAromeffects to promote bond cleavage (117). 36

A.

Crine in DNA active center

Glycosylic HL0 H GIycoyIic 0 a-bofld HN$)a$* HNH \\ a-O.b4tal Anorleric AnOmeliC Eflect Effect p

Figure 3 37

1.3. Other Enzymes with Uracil-DNA Glycosylase Activity in Human Cells

1.3.1. Thymine-DNA Glycosylase

The hymine-NA glycosylase (TDG) activity that removed a thymine residue from a G.T mismatched base pair in synthetic heteroduplex DNA was first detected in HeLa cell nuclear extracts (129,130). Subsequently, the enzyme containing TDG activity was purified to apparent homogeneity from 900 g of HeLa whole cell extracts (131), and shortly afterwards, two human cDNAs encoding TDG were cloned that differed by 100 nucleotides at the 3' untranslated region (132). These TDG cDNAs encode a same polypeptide of 410-amino acids with a calculated molecular mass of 46 kDa (132). Both the native enzyme isolated from HeLa whole cell extracts and the recombinant enzyme from E. coli cell extracts exhibit aberrant mobility during SDS polyacrylamide gel electrophoresis, with an apparent molecular weight of 55 kDa and 60 kDa, respectively (131,132). Using the fluorescence in situ hybridization (FISH) technique and Northern blot analysis, Sard and coworkers mapped the TDG gene to chromosome 12q24. 1 and found the gene was expressed in all human tissues with highest expression in the thymus (133). Biochemical characterization of purified TDG showed that the enzyme could remove thymine residues from duplex DNA containing G'T, C'T and TT mispairs in the order of G'T>> C'T > T.T (131). The enzyme also removed the uracil residues from duplex DNA containing CpG'U, CpCU and CpT.0 base pairs (134). Thekeat for removal uracil residues from duplex DNA containing CpGU, CpCU and CpTU base pairs were 12, 130, 266 times faster than the removal of thymine residues from analogous DNAs containing GT, CT and T'T mispairs (134). Thus, TDG is more active on duplex DNA containing a mismatched uracil than on a mismatched thymine. In addition, the rate of thymine excision by TDG is strongly affected by the sequence

5'-adjacent to the mismatched guanine (134-139). For example, thekeatfor removal of thymine residues from a CpGT mispaired DNA is 40-fold faster than that for TpGT mispaired DNA, and 700-fold faster than that for ApGT mispair DNA (134). Therefore, the TDG may play an important role in removing thymine residues produced through the deamination of 5-methylcytosine at CpG sites of DNA methylation. The low k0ff value (1.8-3.6 x i0 s1) of TDG indicates that after thymine is excised from the DNA, the enzyme remains associated with the abasic DNA (140). The DNA binding assay found that TDG bound tightly (Ki =23 pM) to a duplex DNA containing tetrahydrofliran, an analogue of an AP-site, opposite a guanine (141), and that human apurinic endonuclease (APE 1/HAP 1) significantly enhanced the rate of dissociation of TDG from AP-site DNA, presumably through direct interaction with the bound glycosylase (140). In addition, posttranslational modifications of TDG by SUMO-1 and SUMO-2/3 conjugations reduce the DNA substrate and AP-site binding affinity of the enzyme (142). Beyond G'T and GU mismatches, several studies have demonstrated that TDG can excise thymine residues from duplex DNA containingO6 methylguanineT (135), 6-thioguanineT (138), 2-amino(6-methylamino)purineT (139), and S6-methylthioguanine.T mispairs (134), as well as thymine glycol.G (143) and 3,N"-ethenocytosine base paired with any of the natural bases (77,144). These in vitro results suggest the importance of TDG in the repair of these oxidized pyrimidine bases in vivo. Another biological function of TDG may involve control of transcription. Using a yeast one-hybrid screen assay to search for proteins capable of interacting with the native retinoic acid receptor (RAR) and retinoid X receptor (RXR), Um and co-workers (145) demonstrated that TDG interacted physically with RAR and RXR. The interaction of TDG with RAR/RXR stimulated the binding of RAR/RXR homo- and heterodimers to their cognate responsive elements, and enhanced RARJRXR- dependent transactivation of a reporter gene in transient transfection experiments (145). In addition, Tini and co-workers (146) reported that TDG physically associated with transcriptional coactivators CREB binding protein (CBP)/p300 and stimulated CBP/p300 transcriptional activity in transfected cells. TDG in the TDG-CBP/p300 39 complex served as a substrate for CBPIp300 acetylation, and acetylated TDG subsequently triggered the release of CBP/p300 from DNA ternary complexes and regulated the recruitment of APE 1 to the AP-site (146). Taken together, these observations suggest a potential regulatory role of TDG in transcription regulation.

1.3.2. Single-Stranded Selective Monofunctional Uracil-DNA Glycosylase

By screening in vitro expression of a Xenopus laevis embryo cDNA library using the electrophoretic mobility shift assay and synthetic oligonucleotide probes containing four abasic site analogs, pyrrolidine (Pyr), ring-opened pyrrolidine (roPyr), tetrahydrofuran (THF), and reduced abasic (rAb), Haushalter and coworker (147) identified another uracil-DNA glycosylase (31 kDa), whose activity was insensitive to the inhibition by the PBS1 bacteriophage uracil-DNA glycosylase inhibitor protein (Ugi) (See below). This Ugi-insensitive 31 kDa protein was later named as a ingle- stranded selectiveonofunctional uracil-DNA glycosylase (SMUG 1) because it excised uracil residues preferentially (60-fold) from single-stranded DNA compared to double-stranded DNA (147). The human SMLJG1 (hSMUG1) cDNA was further obtained by searching a human expressed sequence tag (EST) database using Xenopus Smug! as a query (147). The hSMUG1 cDNA encodes a 270-amino acid protein with a calculated molecular mass of -30 kDa, and hSMUG1 shares 60 % (163 out of 270) identity and 70 % similarity (191 out of 270) with the Xenopus SMUG1 (147). hSMUG1 gene was later mapped to chromosome 12q13.1-14 by genomic sequence analysis (147). The expression of the hSMUG1 gene is positively regulated by the transcription factor NFIICTF, and the steady-state levels of hSMUG1 transcript are about 10-fold lower than UNG2 transcript (148). Direct comparison of UNG2 and SMUG! glycosylase activities revealed that both enzymes preferentially excised uracil residues from single-stranded DNA, and were stimulated by physiological concentrations of Mg2, at which the activity of UNG2 was 2-3 orders of magnitude higher than that of hSMUG1 (149). SMUG1 40

showed broader substrate specificity than UNG2, and AP endonuclease had a strong stimulatory effect on SMBG1 activity toward uracil-containing double-stranded DNA, apparently due to enhanced dissociation of SMUG1 from AP-sites (149). Furthermore, the results of a fluorescence microscopy study showed that hSMUG1 accumulated within the nucleoli of cultured epithelial cells, while UNG2 was excluded from nucleoli and accumulated in replication foci during S phase (149). Based on these findings, Kavli and co-workers (149) proposed that UNG2 was responsible mainly for both pre-replicative removal of deaminated cytosine and post-replicative removal of misincorporated uracil at the replication fork, and was the major enzyme for the removal of deaminated cytosine outside of replication foci, with hSMUG1 acting as a broad specificity backup. In vivo, hSMUG1 can functionally compensate for an Ungl deletion in the yeast Saccharomyces cerevisiae (150). In vitro, hSMIJG1 also excises uracil derivatives, 5-hydroxymethyluracil, 5-hydroxyuracil, and 5-formyluracil, and may be important for repair these oxidized DNA damages in vivo (151,152).

1.3.3. Methyl-CpG-Binding Protein 4

Methylation at CpG dinucleotides in eukaiyotic genomes is a fundamental epigenetic control of gene expression in vertebrates and plays an essential role in mammalian development (153,154). Effects of DNA methylation are mediated by DNA binding proteins, which bind to symmetrically methylated 5'CpG sequences through the conservedethyl-CpG-indingomain (MBD) (155). By searching the human EST database for genes encoding an MBD-like motif, Hendrich and Bird (156) identified a family of human methyl-CpG binding proteins, MBD2, MBD3, and MBD4. Meanwhile, by using the human mismatch repair (MMR) protein MLH1 as "bait" and a fetal brain cDNA library as "prey" in a yeast two-hybrid screening, Bellacosa and coworkers independently clonedethyl-CpG binding n4onuclease 1 (MED1)gene, and later found out thatMEDJgene was identical to MBD4 gene (157). 41

TheMBD4/MED1gene encodes a 46 kDa protein with 580 amino acids, containing three distinct domains: a conserved N-terminal methyl-CpG-binding domain (MBD), a central region, and a C-terminal catalytic domain with homology to E. coli MutY and Endo III (156,157). In vitro, MBD4 binds to methylated and unmethylated CpG-containing DNA, removes thymine or uracil efficiently from mismatched CpG sites, and binds 5-methylcytosine TpG.GpC mismatches preferentially (156,157). These results suggest that MBD4 may function as a mismatch-specific DNA glycosylase to remove thymine or uracil residues arising from the deamination of methylcytosine or cytosine and minimize mutation at CpG sites. Indeed,MBD4homozygous knockout mice showed a 3-fold increase in C to T transition mutations at CpG sites. Also, when bred onto the APCM background, MBD4'mice showed accelerated tumor formation with CpG to TpG mutations in the APC gene (158). Recent biochemical studies suggest that MBD4 may play a role in the repair of exogenous chemical exposure-induced base lesions. For instances, MBD4 removes 5-fluorouracil (5-FU) from 5-FUG, thymine from 06-methylguanine'T as well as thymine glycol (Tg) from TgG base pairs, and demonstrates a weak excision activity on 3,1V4-ethenocytosine (cC) from cCG base pair (143,159-161). MBD4 may also be involved in MIMR since the protein directly interacts with MLH 1, and transfection of aMBD4mutant lacking the MBD into mismatch repair-proficient SW480 cells was associated with microsatellite instability (157).

1.4. Uracil-DNA Glycosylase Inhibitor Protein

1.4.1.Bacterial PBS1/2 Uracil-DNA Glycosylase Inhibitor

Bacillus subtilis bacteriophage PBS-i and PBS-2 contain uracil in place of thymine residues in their double-stranded DNA genomes (162). Following phage infection, the incorporation of uracil residues into viral DNA is achieved by the expression of several early gene products. Among these are the dTMP 5 '-phosphatase 42

(163), dCTP deaminase (164), dTJMP kinase (165), and dUTPase inhibitor (166). The products of these genes act to increase the intracellular dUTP pooi while depleting the dTTP pool, and thus facilitating the incorporation of dUMP into viral DNA by a phage-induced DNA polymerase. The newly synthesized uracil-containing DNA must be protected from B. subtilis uracil-DNA repair, as this would lead to degradation of the PBS 1/2 genome (167). This task is accomplished by the PBS 1/2-induced uracil- DNA glycosylase inhibitor protein (Ugi) that specifically inactivates the B. subtilis uracil-DNA glycosylase within four minutes of infection (168). The bacteriophage PBS-2 Ugi was subsequently purified and characterized from extracts of PBS-2 infected B. subtilis (169). The amino-acid sequence analysis of Ugi revealed that 21% of the protein's 84 amino acids were acidic, and isoelectric focusing showed the p1 of Ugi to be -.4.2 (170,171). In vitro, Ugi was found to be a heat stable protein with an apparent molecular weight of -'1 8,000 as determined by gel filtration chromatography (169), and to be resistant to sodium dodecyl sulfate (SDS) and/or 8 M urea inactivation (169,172). However, examination of Ugi mobility by SDS- polyacrylamide gel electrophoresis revealed an apparent molecular weight of -3,500 (170). The apparent molecular weight discrepancy was later resolved with the determination of molecular weight of 9,450 by sedimentation equilibrium centrifugation and 9,474 by mass spectrometry (171). Thus, Ugi exists as a homogeneous monomer in solution (171). Subsequently, the ugi gene of PBS-i was cloned and the nucleotide sequence was reported to be identical to the nucleotide sequence of PBS-2 (173). In vitro studies indicated that Ugi specifically inactivates a variety of uracil-DNA glycosylases isolated from diverse biological sources, such as B. subtilis, E. coli, M luteus, S. cerevisiae, M lactucae, rat liver (nuclear and mitochondrial forms), HSV-1 and -2, human placenta, and KB cells (169,170,172,174,175). In vivo, expression of the ugi gene in E. co/i JM101cells resulted in an ung mutator phenotype with -10 fold higher spontaneous mutation frequencies for nalidixic acid and chloramphenicol resistance (172). Also, expression of Ugi in E. co/i cells made the cells more resistant to the lethal effects of 5- 43

fluorodeoxyuridine. Ugi-expressing human glioma U-251 cells also exhibited a 3-fold higher overall spontaneous mutation frequency compared to control cells, due to increased CG to TA transition mutations. However, the growth rate and cell cycle distribution of Ugi-expressing U-251 cells did not differ appreciably from their parental cell counterpart (176).

1.4.2. Mechanism of Uracii-DNA Glycosylase Inhibitor Action

Ugi is a potent inhibitor for E. coli uracil-DNA glycosylase (Ung), as theK1

(0.14jIM)is one-twelfth of the Km of the enzyme (177). In vitro biochemical studies have demonstrated that Ugi from PBS-2 forms a complex with E. coil Ung, which is stable at physiological conditions and is reversed with the treatment of SDS or 8 M urea at 70°C (171). Both Ung and Ugi activities can be recovered upon dissociation of Ung'Ugi complex (171). In addition, the UngUgi complex has a molecular weight (Mr) of-.'35.4 kDa as determined by gel filtration chromatography (170) and analytical

ultracentrifugation, and is consistent with the combination of Ung (Mr25,664 Da) and

Ugi (Mr 9,474 Da) in 1:1 stoichiometry (171). Furthermore, the isoelectric point of Ung.Ugi was determined to be 4.9, which is between Ung (p1 = 6.6) and Ugi (p1= 4.2) (171). These results indicate that the inhibition of uracil-DNA glycosylase activity by Ugi results from the formation of a stable protein-Ugi interaction in which neither protein nor Ugi are modified (171). A two-step mechanism for Ung-Ugi interaction was proposed from kinetic analysis of stopped-flow experiments (178). The first step involved rapid formation of pre-equilibrium Ung-Ugi complex, presumably corresponding to a docking process, in which the optimal alignment of the two proteins is negotiated. If proper alignment was not achieved, a reversible association would transpire. Thus, the Ung'Ugi precomplex formation was considered reversible, and was distinguished by the dissociation constantKd= 1.3 .tM (178). Alternatively, if proper alignment was achieved, then the second step of the mechanism involved final irreversible complex formation was quickly followed. The formation of the final irreversible "locked" complex involves isomerization of the docked proteins, and was characterized by the rate constant k =

195sec1(178). The K, of Ugi was determined to be 0.14 p.M, which is one-twelfth of theKmof Ung (170). Further investigation of the mode of Ugi inhibition found that binding of Ugi to Ung prevented Ung from binding to uracil-DNA. Ung bound to DNA rapidly bound Ugi and became refractory to DNA binding (171,178). These results suggested that Ugi bound at or near the DNA binding site of Ung. Indeed, results from protein-DNA cross-linking experiments showed that Ung cross-linked to dT20 oligonucleotide DNA was refractory to Ugi binding, and the addition of Ugi to the protein-DNA cross- linking reaction eliminated productive cross-linking of Ung todT20oligonucleotide DNA (179). Thus, these results further suggested that the negatively charged amino acids of Ugi resembled the phosphodiester backbone of DNA. To investigate the role of carboxylic acid amino acid residues in Ugi that may participate in Ung binding, chemical modification of Ugi was performed by treatment with water soluble carbodiimide (EDC) and glycine ethyl ester (GEE) (180). Modification of Ugi protein by EDC/GEE resulted in the selective adduction of specific glutamic acid residues (G1u28 and Glu3l) that abolished the UngUgi interaction (180). Mutational analysis of seven negatively charged amino acid residues (E201, E27A, E28L, E3OL, E31L, D61G, and E78V) of Ugi revealed that each mutant Ugi protein remained capable of associating with Ung and forming a Ung.Ugi complex. However, the stability and reversibility of the complex was perturbed by some amino acid substitutions (181). Thus, the negatively charged amino acid residues of Ugi maymimicthe DNA phosphate backbone and act collectively to facilitate stable Ung'Ugi complex formation. Structural studies of Ugi and Ugi in complex with uracil-DNA glycosylase from herpes simplex virus type-i (HSV-i) (173),E. colE(114,182), and human (183), have provided the molecular basis for the enzyme-inhibitor interaction. As determined by solution state multidimensional nuclear magnetic resonance, the secondary 45

structure of the B. subtilis bacteriophage PBS-2 Ugi protein consisted of five anti- parallel 13-strands and two a-helices, and had six loop or turn regions which contained 7 of the 18 acidic residues (182). These acidic residues were brought together in the tertiary structure to form an "electrostatic knob" that extendedon one face of the protein (184). Analysis of the crystal structure of Ugi in complex with thecore catalytic domain of human uracil-DNA glycosylase (UNG*) revealed that the electrostatic knob interacted with conserved UNG* active site residuesover a large interface, and exhibited electrostatic and shape complementarity to theenzyme (183). In this UNG*.Ugi complex, Ugi was found to insert the 131 beta-sheet into the active site DNA binding groove of Ung; however, the inserted 131 sheet did not directly interact with the uracil-specificity pocket (183). Instead, a cluster of hydrophobic residues in a pocket between the 131 sheet and a2 loop of Ugi enveloped Leu272, which protruded from a sequence-conserved UNG* loop (183). Comparison of UNG* and Ung bound to Ugi with free Ugi revealed that Ugi underwent a conformational change centered at the 131 and 132 sheets upon binding uracil-DNA glycosylase (114,183). The extension of interactions between 131 and 132 was referred to as a 13-zipper. The largest conformational difference was localized around G1u20, which was located at the edge of 131 sheet. The movement of the13- zipper positions G1u20 side-chain and the Gln 19 carbonylgroup to make important interactions with UNG*IUng(1 14,183). Comparison of the structure of UNG* bound to Ugi and UNG* bound to DNA revealed the basis of the enzyme's high affinity for Ugi. Overlays between Ugi and the DNA phosphate backbone from the ten base-pair DNA product complex with UNG* revealed that Ugi bound to theopen (unbound) structure of UNG* (183). The overall size and shape of Ugi corresponded very well with the size and shape of the kinked DNA, and every carboxylate side-chain in Ugi was superimposable on the DNA phosphate backbone (114). These results suggested that Ugi appeared to be a transition-state mimic of flipped-out DNA (114). 46

1.5. Uracil-Initiated Base Excision Repair Pathways in Eukaryotes

Uracil-initiated base excision repair (BER) pathways in eukaryotes are a multienzyme repair process involving uracil excision, phosphodiester bond incision, deoxyribose phosphate removal, DNA synthesis, and DNA ligation (57) (185). The first step of the pathway is accomplished by a uracil-DNA glycosylase that hydrolyzes the Ni-C 1'-glycosylic bond linking the uracil base and deoxyribose, and generatesa free uracil base and an apyrimidinic (AP) site in DNA (57). Following uracil excision, the AP-site is processed by a Class II AP endonuclease, which hydrolyzes the phosphodiester bond on the 5'-side of the AP-site (185). In human, the major human AP endonuclease, named APE!, HAP1 or APEX, is a homolog of theE. coliXth protein (Exonuclease III, Exo III), and was originally discovered as an AP endonuclease and a reduction-oxidation regulator of the DNA binding domain for c- Fos and c-Jun transcription factors (186) (187) (188). Unlike APE 1, the primary AP endonuclease in yeast, termed APN1, is a homolog of E. co/iNfo protein (Endonuclease IV,Endo IV)(189). Both APE1 and APN1 incise the phosphodiester bond 5' to the AP-site through a metal ion-mediated mechanism, leaving a 3 '-hydroxyl and a 5'-deoxyribose phosphate (dRP) group, which is subsequently removed by a deoxyribophosphodiesterase (dRPase) activity (190,191). As a result, the action of APE1 and dRPase on AP-site generates one nucleotide gap in the DNA. The specific activity of APE1 endonuclease is influenced by divalent cations as follows:Mg2> Mn2 >>Ni >Zn2>> Ca2 K (190). In addition to its action on AP-sites in double-stranded DNA, a recent study showed that APE! also acted on AP-sites in single-stranded DNA, but with approximately 20-fold less activity (192). APE1 also possesses a 3'-repair diesterase activity (phosphodiesterase), which was reported to be 100-200-fold lower than its AP endonuclease activity (193). The 3'-phophodiesterase of APE1 is capable of removing a variety of 3'-blocking groups including 3'-4-hydroxy-2-pentenal phosphate and 3'- phosphoglycolate, which must be removed to support DNA repair synthesis and 47

ligation (194,195). Moreover, APE1 displays a minor 3'-5' exonuclease activity for duplex DNA, which is less than 0.03% of its endonuclease activity (196). This exonuclease activity is specific for the 3 '-termini of internal nicks and gaps in duplex DNA, and is enhanced by 3'-mismatched base pairs (197). Recently, Chou and Cheng (198) suggested that the 3' to 5' exonuclease activity of APE 1 may provide a proofreading activity for DNA polymerase f3 (POL ), and thus increase the fidelity of base excision repair. In vivo, APE1 is essential for early embryonic development in mice. The homozygous Apeldeficient mice died during embryonic development following blastocyst formation, shortly after the time of implantation (199). Finally, the human APE2, which differs from APE1 at the N- and C-terminal ends, was recently identified as the second AP endonuclease in human cells with homology to the E. coli ExoIII (200). Unlike APE 1, APE2 exhibits a weak Class II AP endonuclease and 3'-repair activities, and can only partially complement the repair defects in AP endonuclease-deficient bacteria and yeast (200). APE2 contains a putative mitochondria localization sequence, and was shown to localize in both nuclei and mitochondria (201). Moreover, APE2 localized to foci in the nuclei, and co- localized with proliferating cell nuclear antigen (PCNA) in some of the foci (201). APE2 contained a putative PCNA binding motif, which was shown to be functional in immunoprecipitation experiments (201). The role of APE2 in DNA repair remains to be established. After phosphodiester bond incision by a class II AP endonuclease in the uracil- initiated BER pathway, the following steps consist of: 1) the removal of 5'- deoxyribose-5-phosphate (5'-dRP) group; 2) gap filling by DNA polymerase; and 3) DNA ligation. Repair of the gap in duplex DNA has been shown to proceed via a short patch or a long patch repair pathway in eukaryotic systems (202,203). The difference between the two pathways is the size of the DNA repair synthesis patch: either a single nucleotide replacement (short patch), or 2 to8-10 nucleotides (long patch) (203-205). Short patch repair of a one-nucleotide gap has been shown experimentally to be dependent on DNA polymerase f3 (POL J3) (205-208). The involvement of POL j3 in 48

uracil-initiated BER was first evaluated in anin vitrorepair assay using bovine testis nuclear extract and a synthetic duplex 51 -mer DNA substrate containinga single GU mispair (205). When POL was depleted by neutralizing polyclonal antibodies from the nuclear extract, the GU repair activitywas inhibited. However, when aphidicolin, an inhibitor of DNA polymerases a, ö, and, wasadded to the repair reaction, no inhibition was observed (209). Supplementation with purified POL13, but not POL a, or, toDNA polymerase depleted fractions of the nuclear extract restored GU repair activity. Similarly, when double stranded synthetic oligonucleotides containing

an abasic site were used to examine the BERin vitro,short patch BER was observed to be POL f3dependent (204). The requirement of POLin short-patch BER was further strengthened by the report that extracts of the mouse POL f3embryonic fibroblast cell line were defective in uracil-initiated base-excision repair (210). Later

studies established that the function of POL13was to catalyze the release of 5'-dRP residues from incised AP-sites and perform BER-related DNA repair synthesis

(211,212). Thus, it seemed that POL13was the main DNA polymerase responsible for the removal of 5'-dRP residues and one-nucleotide gap-filling in short-patch BER. Several reports indicate that other DNA polymerases are also involved in BER. Using a circular plasmid DNA containinga single abasic site, Fortini and co-workers (213) demonstrated that the POL f3-deficient extracts were able to perform short patch repair on the circular DNA template; however, the extent of short patch repair in the POL 13-deficient extracts was significantly reducedas compared to POL f3-proficient extracts (213). Also, when the repair assay was performed using subfractions containing either POL or POLEfrom the extracts of POL 13-deleted mouse fibroblasts, a single nucleotide repair patch was detected. The rate of the repair by

POL 6 or POL was significantly reduced as compared to that by POL13(214). In contrast, Bennettet al.(215) reported that the efficiency of uracil-initiated BER in POL 13-proficient and -deficient extracts was similar in in vitro repair assays that utilized a covalently-closed, circular Ml 3-DNA substrate containing a site-specific

U'G mismatch. A recentstudy byParlantiet al.(216) reported that short-patch repair synthesis occurred even in the absence of POL f3,and c. Interestingly, POL X was not observed to stimulate this "back-up" DNA synthesis (216). Taken together, these reports indicate that POLf3is not the only DNA polymerase that can perform short patch repair in mammalian cells. POL ö and POLEalso participate in short patch BER repair synthesis. It is possible that other DNA polymerases like POL i., ?, and, in mitochondria, POL y, may be involved in short-patch BER, since these DNA polymerases have been shown to possess an intrinsic dRPase activity (2 17-219). A

in vitro recent study indicates that POLi can substitute for POL1 in BER reactions (217). The role of POL i, X, and ' in BER remains to be established.

POL13,an X family DNA polymerase [the classification of human DNA polymerases is summarized in Table 1] is a 39 kDa protein containing an 8-kDa amino-terminal domain, a protease sensitive hinge region, and a 31 -kDa carboxyl- terminal domain (220). The 8-kDa domain (or lyase domain) possesses a dRP lyase activity that removes the 5 '-dRP moiety created by class II AP endonuclease cleavage of the AP-site intermediate created during base excision repair (211). This dRP lyase activity plays an important role in protecting cells from DNA damage-induced cytotoxicityinvivo (221). The 8-kDa domain also possesses a single-stranded DNA binding activity and binds strongly to the 5'-phosphate group in gapped DNA (222). This single-stranded DNA binding activity of 8-kDa domain directs the holoenzyme to short nucleotide gaps (

synthesis (222). Thus, the processive DNA synthesis of POL13for a gap of six nucleotides or less is modulated by the 8-kDa domain of the enzyme. The 31 -kDa domain (or polymerase domain) performs the DNA polymerization activity (211,220). The acid side chains (Aspl9O, Asp192, and Asp256) of the 31-kDa domain bind two catalytically importantMg2ions, which facilitate phosphodiester bond formation by coordinating the triphosphate of the incoming nucleoside to the 3' -hydroxyl group on i]

the primer terminus (223). A nucleotide-induced fit mechanism has been proposed for POLto perform polymerization reaction as follows: 1) dNTP binding to the polymerase-DNA complex orients the deoxynucleoside triphosphate within the active site; 2) an enzyme conformational change induces the alignment of the catalytic groups for catalysis and closes the C-terminal subdomain around the correct incoming dNTP and its complementary template base (223). Since POLlacks an intrinsic exonuclease (or proofreading) activity, the enzyme performs relatively low fidelity DNA synthesis (224). The polymerase activity of POL f3 is more rapid than the dRP lyase activity; thus, the single nucleotide gap in the duplex DNA is usually filled prior to removal of the 5'-dRP moiety during BER (225). In contrast, inefficient removal of the 5'-dRP moiety by POL J3 causes the strand displacement DNA synthesis by the enzyme, and therefore favors long-patch BER, where the 5'-dRP moiety is removed as part of a single-stranded DNA flap by flap endonuclease 1 (FEN1) (226). This result implies that POL J3 also participates in long-patch BER. Indeed, by analyzing products of long patch excision generated during BER of a uracil-containing DNA substrate in

human lymphoid cell extracts, Dianov et al. (227) showed that POL1 had an essential role in long patch BER by conducting strand displacement DNA synthesis and controlling the size of the excised single stranded DNA flap. The implication of POL j3 in long patch BER was further demonstrated by the ability of both POL 3 andto complete long patch repair synthesis in an in vitro reconstituted repair system using purified proteins (204). The long patch BER pathway has been observed in several in vitro DNA repair system using different eukaryotic cell-free extracts, and the size of the repair patch in different repair systems varies. Using Xenopus laevis oocyte extract and a plasmid DNA substrate containing a site-specific tetrahydrofuran (THF) residue, an AP-site analogue, tostudyrepair of AP-sites, Matsumoto and Bogenhagen (228,229) reported that most repair events involved an ATP-dependent incorporation of no more than four nucleotides. In a separatestudyby Frosina et al. (230) that used Chinese hamster ovary cell extracts and a circular plasmid containing multiple natural abasic sites or 51 methoxyamine (MX)-modified AP-sites, a 7-nucleotide repair patch, on average, was observed for the AP-site DNA substrate, and a lO-nucleotide patch for the MX-AP- site DNA substrate (230). Bennett et al. (215) studied in vitro uracil-initiated BER using a circular substrate and POL 3-proficient and -deficient mouse embryonic fibroblast cell-free extracts and observed that greater than 50% of all repair patches were 2-8 nucleotides in length in both cell extracts. In yeast, it was reported that approximately 50 % of the DNA repair synthesis events during uracil-initiated BER resulted in a one-nucleotide repair patch, whereas repair patches of 2, 3, and 5 nucleotides accounted for 25 %, 13 %, and 5 % of the repair events, respectively (231). Repair patches 2-8 nucleotide in length were also observed during in vitro uracil-initiated BER reactions using human colon adenocarcinoma LoVo cell-free extracts (232). The differences in the length of repair patch may be due to different assay conditions, and to differential extraction of repair proteins in the cell-free extract.

Long patch base-excision repair (LP-BER) involving POL J3,or,and 1ap structure-specificdonuclease 1 (FEN1), is PCNA-dependent. Utilizing fractionated Xenopus laevis ovarian extracts and a plasniid DNA containing a tetrahydrofiiran (THF) residue to study the mechanism of repair of AP-sites, Matsumoto and Bogenhagen first reported that the LP-BER of THF sites was dependent on proliferating cell nuclear antigen (PCNA) and POL ö (229). Also, PCNA-dependent DNA repair synthesis by POL c during uracil-initiated BER was observed in Saccharomyces cerevisiae cell extracts (233). PCNA was found to interact directly with POL 13 and to enhance POL 13-dependent LP-BER of reduced AP-sites (204,234). The requirement of PCNA in long-patch BER was further strengthened by the report that inhibition of PCNA by an anti-PCNA antibody in Chinese Hamster Ovary cell extracts strongly inhibited LP-BER of AP-sites (203). The role of PCNA in LP-BER may be two-fold. First, PCNA may be required to tether POL,POL c, and perhaps even POL 13, to the DNA template for repair synthesis. Second, PCNA interaction with FENI may stimulate cleavage of the displaced 5'-strand (203,214,235,236). The 52

function of PCNA in first aspect in LP-BER is supported by the observation that inhibition of PCNA-dependent DNA replication by p2l" inhibits PCNA stimulation of LP-BER in vitro (237). Similarly, addition of p21 peptide or protein in mouse embryonic fibroblast cell extracts reduces the amount of BER DNA synthesis by 72% and the frequency of LP-BER by 3646% (215). p21, an inhibitor of cyclin- dependent kinase, shares a consensus PCNA-binding motif with POL 6, FEN1 and DNA ligase I (LIG I), and binds to the same region of PCNA (238). Since the binding affmity of p21 to PCNA is higher than that of POL 6, FEN1 or LIG 1(239), binding of p21 to PCNA disrupts PCNA-directed stimulation of POL 6, FEN1 or LIG i in long- patch BER (237). The function of PCNA in facilitating FEN1-mediated DNA cleavage is demonstrated by the observation that a PCNA mutant that does not bind to FEN1 is unable to stimulate FEN1 excision activity during LP-BER (236). It has been established that the PCNA-FEN1 interaction enhances FEN1 binding stability and increases FEN1 nuclease activity (237). FEN1 is a Mg2tdependent and structure-specific endo/exonuclease that removes branched DNA structures with an overhanging single-stranded 5'-unannealed flap, and degrades nucleotides from a nick or a gap DNA progressively (240,241). FEN1 is an evolutionarily conserved component of DNA replication from archaebacteria to humans (242). It processes Okazaki fragments during replication by removing the last primer ribonucleotide on the lagging strand and cleaves a 5-prime flap that may result from strand displacement during replication (243-245). In yeast Saccharomyces cerevisiae, deletion of Rad27, the homolog of human FEN1, results in genome instability and mutagen sensitivity (246). Similarly, a homozygous Feni mutation in mice leads to premature embryonic lethality, while mice that are heterozygous for Feni are viable and show a mild tumor predisposition (247). In addition, defective FEN1 activity in T24 human bladder carcinoma cells is associated with prolonged S Phase delay and impaired DNA repair (248). Reconstitution of the BER pathway with purified human proteins showed that the core BER reaction is modulated by many proteins, such as poly(ADP-ribose) 53 polymerase-1 (PARP-1) and replication protein A (RPA). The involvement of PARP-1 in BER was investigated using PARP-1-deficient mouse embryonic fibroblast cell extracts to repair a single abasic site located in a double-stranded circular plasmid substrate (249,250). As reported, PARP-1-deficient cell extracts were about half as efficient as wild-type cells at the polymerization step of the short-patch repair synthesis, but were highly inefficient at the long-patch repair (249,250). The DNA repair efficiency of cell extracts from mouse embryonic fibroblasts lacking both

PARP-1 and POL13was significantly impaired. Recent studies suggest that PARP-1

not only physically interacts with POL13,but also competes with APE1 binding to 5'- dRP-containing BER intermediate and thus stimulates strand displacement DNA synthesis by POL 13(249,251,252). APE1, like PARP-1, also physically interacts with

f3 in vitro, POL but the interaction ofAPE1 and POL f3 is thought to recruit POL13to the incised AP-site and stimulate 5'-dRP excision rather than strand displace DNA synthesis by POL 13(253,254). The participation of RPA in BER was supported by two observations. First, RPA was reported to stimulate S.cerevisiaeRTH1 (yeast FEN1) nuclease activity (255). Second, RPA was observed to interact with a 20-mer peptide corresponding to the N-terminal region of human UNG2 (256). RPA is a heterotrimeric single-stranded DNA binding protein that is required for DNA replication (257) and nucleotide excision repair (258). Results from fluorescence microscopy and structural studies indicate that RPA and UNG2 co-localized in replication foci and physically associated with each other through the interaction between the C-terminal region of the RPA 32 subunit and N-terminal region of UNG2 (259). It is possible that the RPA-UNG2 interaction might target RPA binding specifically to sites of base damage and recruit the down-stream BER proteins. Results frominvifro BER assays have confirmed that RPA stimulates long-patch BER of AP- sites (260,261). Moreover, recent biochemical results suggest that the mechanism of RPA stimulation in long-patch BER is through the enhancement (-46 fold) of DNA ligase I activity by RPA (262). 54

The -ay repair cross-complementing groupA.(XRCC1) protein is also involved in BER. The XRCC1 gene was originally cloned by its ability to complement a mutant cell line (EM9), which had a 10-fold hypersensitivity to alkylating agents such as ethylmethane sulphonate, and a less than two-fold sensitivity to X-rays (263). An early study showed that XRCCI -deficient EM9 cells exhibited reduced ability to complete the short-patch, but not long-patch BER in vitro, and the defect could be complemented by addition of exogenous recombinant LIG 3a or T4 DNA ligase. These results strongly suggested that the repair defect in BER in XRCC1-deficient cells was stalled at the point of ligation (264). In addition, it was known that levels of LIG 3a polypeptide and activity were reduced -.5 fold in cells lacking XRCC1, and inhibitors of the proteasome, lactacystin and MG- 132, increased LIG 3a protein levels in EM9 cells (265). Even though the level of LIG3a was increased with proteasome inhibitors, deletion or mutation of the BRCT domain in XRCC1 that interacted with LIG 3a disrupted the ability ofXRCC1 to facilitate repair of single-stranded breaks in cells following exposure to alkylating agents (265,266). Disruption of Xrccl in mice resulted in early embryonic lethality (E6.5) (267), and this lethal effect can be rescued by expression of low levels (< 10 %) of Xrccl transgene in mice (268). Taken together, these results suggest that XRCC 1 may be required for the stabilization of cellular LIG 3a and prevention of its degradation by the proteasome, and, at the same time, targeting LIG 3a to sites of strand breakage in BER. XRCC1 may also serve as a scaffolding protein to facilitate the interaction between APE1 and POL 3 during BER (206,269). XRCC1 was found to interact physically with APE1 (269), POL3 (206), and DNA ligase lila (LIG 3a) (270). The XRCC1-POLinteraction is required for efficient base excision repair (271). Other proteins, such as p53 (272) and Werner syndrome protein (WRN) (273,274), may also be involved in modulation of BER efficiency. However, the function of these proteins in BER is still not clear. Following removal of the 5'-dRP moiety and DNA repair synthesis, DNA ligase I (LIG 1) or DNA ligase III (LIG3) performs the last step in BER, ligation of the 3'-OH-5'-PO4nick to seal the DNA phosphate backbone and restore the integrity of 55

the DNA strand. LIG1 has been identified as a component of a high molecular weight replication complex in HeLa cells (275), and is required for the efficient joining of

Okazaki fragments in reconstitution of SV4O DNA replication reactioninvitro (245). LIG3 was originally identified as a protein partner that bound specifically toXRCC1 (270). The human LIG 3 gene encodes two nuclear isoforms, denoted LIG3a and LIG3f3, which differ in the length of their C-termini (276-278). While LIG3a associates with XRCC1 through its C-terminalBRCTdomain, LIG3 lacking aBRCT domain does not (277,279). The requirement of LIG1 in short-patchBERwas first reported in the study using a bovine testis crude nuclear extract to conduct uracil- initiatedBER in vitro(205). In the study, a multiprotein complex that catalyzed the repair of uracil-containing DNA substrate was partially purified by affinity chromatography from a bovine testis crude nuclear extract. LIG1 was found in this complex. Subsequent studies showed that the N-terminal domain of LIG1 interacted

directly with the 8-kDa domain of POLI (280,281) and with PCNA (282), and demonstrated that LIG 1 was involved in long-patchBER(283,284). Unlike LIG1, LIG3 is thought to participate in short-patchBERonly, since the CHO EM-Cl 1 cell- free extracts with reduced LIG3 activity were defective in short-patchBERbut proficient in long-patchBER(264). The effect of LIG1 deficiency in short and long BERhas not been determined.

Table 1 Classification of Human DNA Polymerases (285)

Greek dame HUGO name Clam Other names Proposed main function a (slpltnit POLA B POLl DNA replication p(beta) PCILB X Base excision repair ppamniat POLO A MWI Mitochonirial rlicatioti ftdelta) POLDI. B POLl DNA replication epsilon POLE B POL2 DNA replication (zeta) POLl B REV? 13ypass synthesis s tetal POLH V RADIO, XPV 13ypass synthesis (I theta) POLQ A inug08. eta DNA repair st itt POT 1 BAD 11(13 B pa tnthe a a (kappat POLK V DirnDl, theta Bypass synthesis A (lambda) POLL X POL4. beto2 Bose excision repair Jo mu PoI.itr X Non-hornoligono. end joining a (sigma POLE X TRF4. kappa Sister chromatid cohesion REV1L Y REVI Bypass synthesis TDT X Antigen receptor diversity 56

1.6. Research Objectives

Mutational studies indicated that the leucine-loop residues His268, Ser270 and Leu272 were critical to UNG* activity. His268 is involved in active site chemistry, whereas Ser270 is required for DNA phosphate interactions (111). Substitution of Leu272 with alanine altered DNA binding affinity as well as catalytic activity on both UG- and UA-containing DNAs (116). Although the role of the leucine-loop residues His268, Ser270, and Leu272 in IJNG* catalysis has been examined (111,116), the function of Arg276, located at the C-terminal end of the leucine-loop, is unknown. As illustrated in Figure 12B, theand r nitrogens of the Arg276 guanidinium side chain are structurally competent to interact with the third DNA phosphate 3' to the uracil residue, and to participate in water-bridged hydrogen bonds with the N3 of the purine adjacent to the uracil nucleotide as well as with the carbonyl group of Leu272 (115- 117). These observations suggested that the function of Arg276 might involve protein- DNA interactions and stabilization of the leucine-loop and Leu272 side chain, either before or after it is inserted into the DNA minor groove. As a first step in elucidating the role of Arg276 in protein-DNA interactions, random oligonucleotide-directed codon-specific mutagenesis was conducted on the Arg276 codon of UNG. It was anticipated that biochemical, structural, and kinetic analysis of the Arg276 mutant proteins would provide important insights into the interaction of TJNG uracil-DNA. Studies were also undertaken to assess the effects of NaC1 andMgC12on the activity and uracil binding-associated conformational change of full-length UNG2. Subsequently, the similarities and differences between UNG2 and UNG were analyzed. The results provided in this dissertation represent the first evidence that MgC12influences the UNG2 conformational change upon binding to uracil-containing

DNA. Finally, studies were conducted to determine whetherE.co/i Ung and human UNG utilized a "Push-Pull" or "Pull-Push" mechanism to flip the uracil nucleotide. These results represent the first report to determine the biochemical and kinetic similarities and differences betweenE.co/i and human UNG. 57

2. MATERIALS AND EXPERIMENTAL PROCEDURES

2.1. Materials

2.1.1. Chemicals

Chioramphenicol, kanamycin, N-(2-hydroxyethyl)piperazine-N'-(2- ethanesulfonic acid) (HEPES), EDTA, boric acid, ammonium hydroxide, ammonium acetate, 3-(cyclohexylamino)- 1 -propanesulfonic acid (CAPS), dimethylsulfoxide (DMSO), phenylmethylsulfonyl fluoride (PMSF), magnesium chloride, potassium chloride, potassium phosphate, potassium phosphate (dibasic), sodium chloride, sodium phosphate, sodium phosphate (dibasic), rifampicin, streptomycin sulfate, formamide, bovine serum albumin (BSA), -mercaptoethanol, Tween-20, adenosine triphosphate (ATP), thiamine, ammonium hydroxide (30%)were obtained from Sigma. Isopropyl-f3-thiogalactopyranoside (IPTG), 5-bromo-4-chloro-3-indolyl -D- galactopyranoside (X-Gal), dithiothreitol (DTT), 1 kb DNA ladder, sodium dodecyl sulfate, glycine, phenol, sodium dodecyl sulfate (SDS), sucrose, ammonium sulfate, and agarose were purchased form Invitrogen. Benzamidine hydrochloride was from Calbiochem. Dextrose, urea, sulfuric acid, monobasic and dibasic potassium phosphate, 1-butanol, iso-amyl alcohol, glycerol, Tris (Tris-base), and dimethylformamide were obtained from J.T. Baker. Distilled glycerol was bought from EM Science. Blue dextran was purchased from Amersham Biosciences. Fischer was the source of ampicillin, tetracycline, methanol, toluene, acetic acid, formic acid, hydrochloric acid, 2-propanol, and trichioroacetic acid. Acrylamide (>99% pure), bis N,N'-methylene-bis-acrylamide, ammonium persulfate, TEMED, bromophenol blue, xylene cyanol FF, Coomassie Brilliant Blue G250, low-range protein marker, kaleidoscope prestained protein standards, and Bradford reaction dye reagent concentrate used for protein assay were obtained from Bio-Rad. Polyvinylidene fluoride (PVDF) membrane was purchased from Schleicher & Schuell. 58

2.1.2. Radioisotopes

[y-32P]ATPwas purchased from PerkinElmer Life Sciences.

2.1.3. Bacterial Media

Yeast extract, tryptone and bacto-agar were purchased from Fisher. LB medium was composed of 0.5 % yeast extract, 1 % tryptone, and 1 % NaC1. 2xYT medium was comprised of 1 % yeast extract, 1.6 % tryptone, and 1 % NaC1. SOC medium contained 0.5 % yeast extract, 2 % tryptone, 0.05 % NaC1, 2.5 mM KC1, 10 mM MgC12, and 20 mM glucose and the pH was adjusted to pH 7.0 with 5 N NaOH. M9 medium contained 1.28 % (w/v) Na2HPO47H20, 0.3 % (w/v) KH2PO4, 0.05% (w/v) NaCI, and 0.1 % (w/v) NH4C1. Following sterilization, Terrific broth consisted of 2.4 % (w/v) yeast extract, 1.2 % (w/v) tryptone, 0.4 % (w/v) glycerol, 17 mM KH2PO4, and 72 mM K2HPO4. M9 medium was adjusted to 2 mM MgSO4, 0.1 mM CaC12, 0.4 % (w/v) glucose, and 10 jig/mi of thiamine from individual sterile stocks. Solid plates and top agar were prepared by the addition of 1.5 % and 0.7 % (w/v) bacto-agar (Difco), respectively, to the appropriate liquid media. Where appropriate, liquid media or agar plates were supplemented with 1% glucose (Dextrose), or with 100 jig/mi ampicillin from a filter-sterilized stock (100 mg/mi), 25 jig/mI of kanamycin from a 25 mg/mi stock, 34 jig/mI of chioramphenicol from a 34 mg/nil stock or 15 jig/mi tetracycline from a 15 mg/mi stock (in 100% ethanol). To detect a- complementation of lacZa, M9 top agar was adjusted to 0.4 mM JPTG and 1 mg/nil X-Gal.

2.1.4. Bacterial Strains

The strains ofE.coli used in this research and their representative genotypes are listed in Table 2.E.coli strains BH156, BH157, and BH158 were provided from 59

A.S. Bhagwat (Wayne State University). E. coli strain BW1067(ung-153::kan thi-1

reiAl spoTl)was a gift from B. Weiss (Emory University) andE.coli JM1O5 and phage Plvira was provided by W. Ream (Oregon State University).E.coli JM1O9

were purchased from Stratagene, and E. coli BLR from Novagen. UsingE. coil

BW1067 as the donor strain and 1M105 as recipient, 1M105(ung::kan),renamed as CY1O, was created using standard phage P1 transduction methods described by Miller

(286). Similarly, usingE.coil B}1157 as the donor strain and JMIO5 as recipient,

JIM1O5(dug::tet),renamed as CYO1, was created by phage P1 transduction.E. coil

CY1O was transduced torecA (L(srl-recA)3O6::Tn10(Tc'))using phage P1 lysate from the donor strain BLR; this strain was called CYlOrec.E. co/i CY1 1 (ung dug) was created by transduction of CY 10 with phage P1 lysate from the donor strain BH157. The characterization of E. coli CY strains was shown in Table 3-4 and Figure

4-6.E.coil CYlOrec and CY1 1 were transformed with pRP (Cm') that contained the arginine AGA/AGG tRNA and proline CCC tRNA genes (Stratagene).

2.1.5. Plasmids and Bacteriophage

Bacteriophage M13mp2 was obtained from T.A. Kunkel (National Institute of Environmental Health Sciences). The expression vector pKK223-3 was obtained from Amersham Biosciences. Plasmid pKK-Dug, an expression vector for double-strand uracil DNA glycosylase (Dug), was constructed as described by Jung and Mosbaugh (78). Plasmid pRIL and pRP were obtained from Stratagene. Plasmid pET22b and pET28a were purchased from Novagen. Plasmid pUNG15 was obtained from American Type Culture Collection (#65269). Plasmid pTrc99A was from Amersham Biosciences. 2.1.6. Chromatographic Resins

Nickel-nitrilotriacetic acid agarose was obtained from QIAGEN. Hydroxyapatite Bio-Gel HTP, Macro-Prep CM Support weak cation exchange resin, Bio-Gel P-4 Bio- Gel (130±40 pm), and AG 501-X8(D) (20-50 mesh, chloride form), AG 1-X8 (100- 200 mesh, chloride form) ion exchange resins, and AG 501-X8(D) mixed bed resin were purchased from Bio-Rad, and poiy r(U) Sepharose, CM-Sephadex, Q Sepharose fast flow strong anion exchange resin, Sephadex G-75, and Sephacryl S-500were from Amersham Biosciences. DEAE-cellulose (DE-52) anion exchange resin and phosphocellulose (P-i 1) cation exchange resin were obtained from Whatman.

2.1.7. Oligonucleotides

Oligonucleotides used for this research are listed below. Commercial synthetic deoxyribonucleotides, 2'-deoxyuridine-25-mer (U-25-mer), gel-purified 2'- deoxypseudouridine-25-mer (dU-25-mer), 2'-deoxyadenine-25-mer (A-25-mer), thymine-25-mer (T-25-mer) and 2-aminopurine-25-mer (2AP-25-mer) were obtained from TriLink Biotechnologies. Oligonucleotides U-30-mer, A-30-mer and PCR primers were made by IDT biotech. Carboxyfluorescein (PAM) 5'-end labeled U-25- mer (5'-FAM-U-25-mer) and T-25-mer (5'-FAM-T-25-mer) were synthesized by MWG Biotech. 1) PCR primers forE. co/i ung gene verification FP: 5'-GCAGTTAAGCTAGGCGGATTG-3' RP: 5'-TGCCATCCGGCATTTCCCC-3' 2) PCR primers for E. co/i dug gene verification P1 -33-mer: 5'-CAGAATTCATGGTTGAGGATATTTTGGCTCCAG-3' P2-3 3-mer: 5'-CCAAGCTTTTATCGCCCACGCACTACCAGCGCC-3' 3) Oligonucleotides for nuclease assay 5'-FAM-U-25-mer: 61

5'-FAM-GGGGCGCGTAUAAGGAAUCGTACC-3' 5'-FAM-T-25-mer: 5 '-FAM-GGGGCGCGTAUAAGGAATFCGTACC-3' A-25-mer: 5'-GGTACGAATTCCTTATACGAGCCCC-3' 4) Oligonucleotides for DNA binding assay

diiiU-25-mer: 5'-GGGGCTCGTAjjAAGGAATTCGTACC-3' 2AP-25-mer: 5'-GGTACGAATTCCTT2APTACGAGCCCC-3' 5) Oligonucleotides for photochemical cross-linking reactions

dMIU-25-mer: 5'-GGGGCTCGTAAAGGAATTCGTACC-3' U-25-mer: 5LGGGGCTCGTAUAAGGAATTCGTACC3' T-25-mer: 5'-GGGGCTCGTATAAGGAATTCGTACC-3' 6) Oligonucleotides for enzyme processivity assay U-30-mer: 5'-GCGTGACGCACTGAUAAGTGAATTCGACCG3' A-30-mer: 5'-CGTCACGCCGGTCGAATTCACTTATCAGTG-3' 7) Oligonucleotides for uracil-DNA glycosylase activityassay 5'-FAM-U-25-mer: 5 '-FAM-GGGGCGCGTAUAAGGAATTCGTACC-3' A-25-mer: 5'-GGTACGAATTCCrFATACGAGCCCC-3' 8) PCR primers for construction of TJNG*/pTrc99A vector Forward primer (RI): 5'-GCGAATTCTTTGGAG AGAGCTGGAAG-3' Reverse primer (113): 62

5'-GCAAGCTTTCACAGCTCCTTCCAGTC-3' 9) PCR primers for construction of UNG/pET-22b vector Forward primer: 5'-GCGAATTCCATCACCATCACCATCACTTTGG- AGAGAGCTGGAAG-3' Reverse primer: 5'-GCAAGCTTTCACAGCTCCTTCCAGTC-3' 10) Oligonucleotides for random mutagenesis FP-3 1 -mer: 5'-CTTTGTCAGTGTATNNNGGGTrCTTTGGATG-3' RP-32-mer: 5'-CATCCAAAGAACCCNNNATACACTGACAAAGG-3' 11) Oligonucleotides for site-directed mutagenesis FP-3 1-mer: 5'-CTTTGTCAGTGTATXXXGGGTTCTTTGGATG-3' Where XXX was TGC for cysteine, CAC for histidine, ATG for methionine, GTA for valine, and TGG for tryptophan substitutions. RP-32-mer: 5'-CATCCAAAGAACCCYYYATACACTGACAAAGG-3' Where YYY was GCA, GTG, CAT, TAC, and CCA for the cysteine, histidine, methionine, valine, and tryptophan substitutions, respectively.

2.1.8. Enzymes

Restriction endonucleasesDpnI,EcoRI,Hindu,Sad, andNdeIwere obtained from New England Biolabs, as were E. coli exonuclease III and Deep Vent DNA polymerase. Pfu Turbo DNA polymerase was purchased from Stratagene. T4 polynucleotide kinase and T4 DNA ligase were obtained from Fermentas. E. coli endonuclease IV (fraction V) was provided by B. Demple (Harvard University). Bacteriophage PBS-2 uracil-DNA glycosylase inhibitor (Ugi) protein (fraction IV) and 63

E. coli Ung (fraction V) were purified as previously described by Sanderson and Mosbaugh (180).

2.1.9. Animals

Two female New Zealand white rabbits (10-week-old) were raised in the animal facility of Laboratory Animal Resource Center at Oregon State University.

Table 2. E. coil strains and genotypes

E. coli Strains Genotype Reference

GM3 1 dcm-6 thr-1 hisG4 leuB6 rpsL ara-14 supE44 lacYl tonA3l tsx-78 galK2 x (287)

BH156 GM31 with ung-1 tyrA::TnlO (79)

BH157 GM31 with mug::TnlO (79)

BH158 GM31 with ung-1 tyrA::TnlO mug::TnlO (79)

JM1O5 F' traD36 lacFA(lacZ)M15 proAB/thi rpsL (Str') endA sbcC hsdR4(rk-mk+) (288) A(lac-proAB)

CY1O JM1O5 with ung::kan This study

O(TCR) CY 1 Orec JM1 05 with ung: :kan A (sri-re cA)306::Tn 1 This study

CYO 1 JM 105 with dug::tet This study

CY1 1 JM1O5 with ung::kan dug: :tet This study

JM1O9 recAl e14(McrA) ii(lac-proAB) thigyrA96 (Nat) endAl hsdRl7 (r mkrelAl supE44/ F traD36 iacP A('iacZ,)M15proAB (288) 64

Table 3. Uracil-DNA glycosylase activity of JM1O5 transductants

Extract Concentration Specific Activity Strain Total Units (mg) (Units/mg)

JM1O5 17.7 359.5 20.3

CY1O(ung) 13.6 4.3 0.3

CYO1 (dug) 14.5 388.7 26.8

CY1I (ungdug 12.7 2.5 0.2 65

Table 4. M13 uracil-DNA phage infection assay of E. coli JM1O5 and CY strains

Number of blue Strain M13 phage dilution plagues JM1O5 10 0 iø- 1 10 34 iO 253 CY1O(ung) 10b0 6 i09 42 108 469 iO >1000 106 >iøøø CYO1 (dug) 106 0 1 io 30 10 240 CY11 (ungdug) 10b0 6 io 69 108 540 io >1000 106 >1000 Figure 4. Validation of E. coli CYO1 and CY11 by cC-excision activity assay. Six sets of standard DNA glycosylase reaction mixtures (10 j.tl) containing double- stranded {32P]34-mer oligonucleotide substrates (10 ñM) with cCG-target residues were prepared as described in Section 2.2.12.2. After adding 4 nM of Dug (lane 2) or 100 jig of cell free extracts from JM1O5, CY1O, CYO1, CY1 1 strain (lanes 3-6, respectively), each reaction (10 p1) was incubated for 30 mm at 30 °C. Control reactions containing DNA substrate without enzyme addition (lane 1) were also performed in an identical way. Following the incubation, 5 j.tl of each reaction was supplemented with 1 unit of Endo 1V (1p1)to cleave the AP-sites. Samples were heated for 3 mm at 70 °C to stop the Endo IV reaction and mixed with 2x sample buffer. Reaction products were analyzed by denaturing 12% polyacrylamide/8.3 M urea gel electrophoresis. The substrate and product bands, designated as S (34-mer) and P (1 5-mer), respectively, are indicated by arrows. The cC-excision activity was not detected in CYO1 or CY1 1 cell free extracts (Lanes 5 and 6, respectively). 67

32p5' 3' G Lane:1 23456

S (34-mer)

P (15-mer)

Figure 4 r4 LlI

Figure 5. Validation of E. coli CY11 by PCR of E. coli ung and dug genes.JM1O5 (lane 2 or 8) and four CY1 1 candidates (lanes 3-6 or 9-12, respectively) are shown. Cell pellets (10 i.il) from 2 ml of overnight culture in 2x YT medium were resuspended in 50 p1 of deionized water and boiled at 100°C for 10 minutes. The cell resuspension was clarified by centrifugation at 14,000 xg for 10 mm at 4 °C. A portion (5 iii) of each supernatant was then used forE, coIl ung or dug PCR reactions. For E. coli ung (lanes 2-6) or dug (lanes 8-12) PCR reactions, reaction mixtures were supplemented with forward and reverse primers (1 p.M) forE. coli ung or dug gene (See Section 2.1.7), 10 unit of Pfu Turbo DNA polymerase, and 10 j.tM dNTPs. Following the PCR reactions, each reaction action mixture (20 p1) was mixed with 4 p1 of 6x DNA loading dye (Fermentas). Each sample(5j.tl) of the PCR reaction product was subsequently analyzed by 1.5% agarose gel electrophoresis. The samples (1 1g) containing a 1-kb DNA ladder were used as standards (lane 1 and 7, respectively). The horizontal arrows indicate the size of reference DNA standards expressed in base pairs (bp). The location of PCR products forE. coli ung and dug gene are indicated by arrows. As shown in lanes 3-6 and 9-12, respectively, no detectable PCR products forE. coli ung or dug gene was found for these four CY1 1 candidates. L

2000 bp 1000 bp ung 500 bp

Figure 5 70

Figure 6. Validation of E. coli CY1O strain lacking recA gene by ultraviolet light exposure. Two sets of five replica plates containing CY1O (upper row) or CYlOrec cells (lower row) were exposed to UV radiation for 0, 5, 10, 15, and 20 minutes, respectively, in TJV Stratalinker 1800 (Stratagene). Cell plates were incubated at 37 °C overnight and then photographed. As shown in the lowerrow, CYlOrec cells are more sensitive for UV radiation. 71

Exposure time (mm)

Figure 6 72

2.2. Experimental Procedures

2.2.1. Preparation of Chromatographic Resins

2.2.1.1. Preparation of DE52, P-il, Sephadex G-75, Bio-Ge1P-4, and Hydroxyapatite Bio-Gel HTP

Sephadex G-75 was equilibrated and stored in UEB buffer (10 mM Hepes- KOH (pH 7.4), 10 mM 2-mercaptoethanol, 1 mM EDTA, 1 M NaCI and 5% (w/v) glycerol) for the purification of Dug. Hydroxyapatite Bio-Gel HTP powder was suspended in 10 volumes of 0.5 M potassium phosphate (pH 7.5). After settling, the buffer was decanted and the resin suspended in 10 volumes of 20 mlvi potassium phosphate (pH 7.5). The resin was defined at least 5 times before storing as a 50% (v/v) slurry in 20 mM potassium phosphate (pH 7.5) at 4°C. Pre-swollen DE52 diethylaminoethyl cellulose was used to prepare DEAE-cellulose resin. The resin was defined at a 1:5 ratio (w/v) and equilibrated in buffer SB (50 mM Tris-acetate (pH 7.0), 10 mM NaC1, 0.5 mM PMSF, 1 mM DTT and 10 % (w/v) glycerol) for the purification of UNG* protein. Bio-Gel P-4 was equilibrated in TE buffer (10 mM Tris-HC1 (pH 8.0), 1 mlvi EDTA) and stored as a 50 % slurry.

2.2.1.2. Preparation of Dowex 1-X8 Ion Exchange Resin

Dowex AG 1-X8 (100-200 mesh, chloride form) was converted to the intermediate hydroxyl counter ion form prior to conversion to the final formate counter ion form for use in the uracil-DNA glycosylase activity assay. Dowex AG 1- X8 was added to 400 ml of fresh 1 M NaOH until the volume was equal to 500 ml. After mixing, the resin was allowed to settle for 20 mm and the 1 M NaOH was decanted by aspiration. This step was repeated until 2000 ml of 1 M NaOH had been used. The resin was transferred to three 150 ml glass filter funnels (60-C) and the remaining 1 M NaOH was filtered from the resin by gravity. The contents of each 73

funnel was washed with 100 ml of 1 M ammonium formate buffer (pH 4.2) followed by washing three times with '333 ml of 10 mM ammonium formate buffer (pH 4.2). The resin was allowed to fully drain between each addition. Finally, Dowex AG 1 -X8 was stored at 4°C in 10 mM ammonium formate buffer (pH 4.2). The 1 M ammonium formate buffer (pH 4.2) was made from 100% formic acid (HCOOH) and 30% ammonium hydroxide (NH4OH).

2.2.1.3. Preparation of Single-stranded DNA Agarose

Calf thymus DNA (Type I, Sigma) was solubilized at 15 mg/mi in 100 ml of 20 mM NaOH and slowly stirred overnight at room temperature. The DNA solution was incubated at 95°C for 15 mm and added to an equal volume of molten 4% agarose equilibrated at 70°C. After mixing thoroughly, the DNA-agarose mixture was poured into an ice cold glass dish (Pyrex, 196 mm x 100 mm) held on ice and allowed to solidify. The solid DNA-agarose mixture was passed twice through a stainless steel sieve (60 mesh) and the resultant gel particles were suspended in 100 ml of resuspension buffer (10 mlvi Tris-HC1 (pH 7.5), 1 mlvi EDTA, 100 mlvi NaC1). Each resuspended gel mixture was placed in a Buchner funnel (17 cm diameter) and washed with approximately 10 L of resuspension buffer at room temperature until theA260 of the filtrate was below 0.02. The single stranded-DNA agarose was stored in the resuspension buffer as a 50 % (v/v) slurry at 4 °C.

2.2.2. Miscellaneous Methods

2.2.2.1. Preparation of Dialysis Tubing

Dialysis tubing (SpectraPor) was cut into lengths of 20 to 30 inches and soaked in 1 % acetic acid solution for 1 h. The soaking solution was decanted and tubing was 74

rinsed with distilledH20prior to boiling in 1 L of 1 % NaHCO3 and 0.1 % EDTA with intermittent stirring. The NaHCO3/EDTA solution was exchanged after it became cloudy and/or yellow. This step was repeated three times. The tubing was rinsed and boiled in distilledH20,rinsed again, and stored at 4°C in 10 mM EDTA (pH 8.0).

2.2.2.2. Preparation of Oligonucleotides

Oligonucleotides U-30-mer and A-30-mer were subjected to further purification by polyacrylamide gel electrophoresis. Samples (300-500 p.1 containing 100-200 iimol oligonucleotide) were combined with native sample buffer to a final concentration of 50 mM Tris-HC1 (pH 6.8), 10 % (w/v) glycerol, and 0.1 % bromophenol blue and loaded onto a nondenaturing 15 % polyacrylamide gels (30 x 40 x 0.16 cm) buffered with 1 x TBE (90 mM Tris, 90 mM boric acid, 2 mM EDTA). Electrophoresis was conducted using TBE buffer at 1000 V until the tracking dye had migrated 25 cm. After electrophoresis, the polyacrylamide gel was placed on top of a sealed TLC plate (Polygram Ccl 300 PEI/UV254) and the oligonucleotide bands were briefly visualized by UV-shadowing. Oligonucleotide bands were cut out from the gel using a clean razor blade and the gel slices were placed into a sample chamber of ELUTRAP electrophoresis chamber (Schleicher & Schuell) containing2 L of 0.5 x concentration of TBE buffer. Oligonucleotides were electroeluted in 0.5 x TBE buffer at 200 V for 2 h in the ELUTRAP electrophoresis chamber (1.2 x 10 cm). Following electroelution, the current was reversed for 30 sec and the DNA solution in the sample trap (-500-1 000 p.1) was transferred into a clean 1.5 ml microcentrifuge tube. The concentration and purity of oligonucleotide was determined by the ratio of the absorbance reading at0D2601280 ,and oligonucleotides were stored at -20°C. 75

2.2.2.3. 5 '-End Phosphoiylation of Oligonucleotides

Oligonucleotides (4 nmol) were 5'-endphosphorylated in reaction mixtures (137.5 j.tl) containing 50mM Tris-HC1 (pH 7.6), 10mM MgC12, 5 mM DTT, 0.1 mM EDTA, 250 xCi [y-32P]ATP (6,000 Cilmmol), and 50 units of T4 polynucleotide kinase. Samples were incubated for 15mmat 37°C and adjusted to 200 .tM ATP using a 5 mM ATP stock. The reaction was incubated an additional 45 miii at 37°C and terminated by adjustment to 10.7 mM EDTA and heated for 10 mm at 70°C. The terminated reaction volume was adjusted to 250 j.d with TE buffer and unreacted [y- 32P]ATPwas removed from the mixture by passage through two consecutive P-4 (Bio- Rad) spin columns containing 1.4 ml of resin and equilibrated in TB buffer. P-4 spin columns were centrifuged in a clinical centrifuge (IEC, setting #4 for 2.5mm).

2.2.2.4. Annealing Reaction of Duplex DNA

The duplex DNA in the study was annealed by heating the oligonucleotides (5 nmol) in buffer C (500 p.1) to 85 °C for 5 mm followed by slow cooling to room temperature. Alternatively, oligonucleotides [32P]U-30-mer and A-30-mer were annealed in a Hybaid PCR Express Thermocycler using the following parameters: 2 miii each at 85, 75, 65, 60, 55, 50, 45, 40, 35, 25, 15, and 5 °C.

2.2.2.5. Protein Concentration Measurements

The protein concentrations of E. coli Ung and Ugi were determined by absorbance spectroscopy using the molar extinction coefficients c280 = 4.2 x 1 liters/molcm (Ung) and28o , =1.2 x 1 liters/molcm (Ugi). The concentration of E. coli cell-free extract, human UNG* and UNG or R276X mutant protein was measured by the Bradford reaction (289) using the Bio-Rad protein assay reagent, which was capable of accurately determining concentrations of 5-15 p.g/ml extract ,I1

protein. Bovine serum albumin (BSA) was used as the protein standard for the assay. The concentration of the BSA standard was determined using the molar extinction

coefficients280 = 0.67 mllmgcm. Standards were diluted from 0-20 jtg/ml in 2.5 tg/m1 increments to generate a standard curve.

2.2.2.6. Rapid Protein Staining of Nondenaturing Polyacrylamide Gels

Proteins resolved by nondenaturing polyacrylamide gel electrophoresis were visualized by staining with Coomassie Brilliant Blue G-250 in 3.5 % HC1O4 as described by Reisner (290). The rapid staining solution was prepared by diluting 100 ml of 70 % HC1O4 into 1900 ml of distilled H20 followed by the addition of 0.8 g of Coomassie Brilliant Blue G-250. The mixture was stirred at room temperature for 1 h and filtered through Whatman No. 1 paper. Nondenaturing polyacrylamide gels were submerged in '-' 500 ml of rapid stain solution and most protein bands were typically visualized after 10 mm of incubation with gentle agitation at room temperature. In some cases, gels were stained overnight. The background color of the gel turned a light amber color while protein bands stained a darker amber hue. After staining, the gels were placed in 5 % (v/v) acetic acid (-P500 ml) which helped increase the sensitivity of detection by about three-fold. Under these conditions, the background color of the gel turned a light blue while protein bands stained a dark blue. Several changes of the 5 % acetic acid wash greatly reduced the light blue background color.

2.2.2.7. Preparation of Competent Cells

Terrific Broth (291) (50 ml) containing kanamycin (25 j.tg/ml) and chioramphenicol (34 tg/m1) was inoculated with an E. coli CYlOrec/pRP colony and incubated with shaking (250 rpm) overnight at 25 °C. Cells were harvested in early log phase(OD600 = 0.4), incubated on ice for 10mm,harvested by centrifugation, and processed according to the standard protocol of the Z-Competent Transformation 77

Kit (Zymo Research). Aliquots (100 1.d) of the competent cell preparationwere stored at -80 °C.

2.2.2.8. Transfection of Competent E. coil Cells by Electroporation

Electroporation was conducted using a Gene Pulser electroporation system (Bio-Rad) and 0.2 cm gapped electroporation cuvettes. Toa prechilled 1.5 ml microcentrifuge tube, competent E. coil cells (- 50 !.tl) were mixed with 0.5-2.5 tl of transfection DNA and held on ice for 1-2 mm. The DNA/cell mixture was transferred to a cold electroporation cuvette and placed into the chamber with tapping to eliminate any air bubbles. A single pulse set at 25 j.tF capacitance, 2.5 kV, and 200-400 was applied. The time constant of the pulse was -P9.3 ms. Transfected cells were recovered in 1 ml ofSocmedium and then incubated at 37 °C for 30 mm. The recovery cell solution was serially diluted into LB or 2xYT medium (1/102 to 1/10) and a sample (250 j.tl) were spread on prewarmed 2xYT agar plates containing 1% glucose and necessary antibiotics.

2.2.3. Electrophoresis

2.2.3.1. Sodium Dodecyi Sulfate Polyacriamide Slab Gel Electrophoresis

Sodium dodecyl sulfate polyacrylamide slab gel electrophoresis was performed similarly to that described by Laemmli (292). Slab gels contained a resolving gel (13 cm) composed of various concentrations of acrylamide and N,N'- methylenebis(acrylamide) (typically 20 %:0.53 %, 12.5 %:0.33 %, or 10 %:0.27 %) from a stock solution of 30 % acrylamide:0.8 % N,N'-methylenebis(acrylamide) (37:1 ratio), 0.1 % SDS, and 375 mM Tris-HC1 (pH 8.8). The resolving gelwas polymerized by adjustment to 0.03 % (w/v) ammonium persulfate and 0.07 5 % (vlv) TEMED. The stacking gel (1 cm) contained 3 % acrylamide, 0.08 % N,N-methylenebis(acrylamide), 78

0.1 % SDS, and 125 mM Tris-HC1 (pH 6.8). The stacking gel was polymerized by adjustment to 0.10 % (wlv) ammonium persulfate and 0.10 % (v/v) TEMED. Protein samples were mixed with an equal volume of tracking dye buffer (50 mM Tris-HC1 (pH 6.8), 1 % SDS, 143 mlvi 2-mercaptoethanol, 10 % (wlv) glycerol, 0.04 % bromophenol blue) and heated at 100°C for 10 mm before being loaded onto the gel. Electrophoresis was conducted at room temperature at 100 V until the tracking dye reached the resolving gel and then the voltage was increased to 200 V until the tracking dye was -.2 cm from the bottom of the gel. The running buffer contained 25 mM Trizma base, 192 mM glycine, and 0.1 % SDS. Gels were fixed in a solution containing 10 % acetic acid and 50 % methanol and protein bands were visualized by staining in a solution of 10 % acetic acid, 50 % methanol, and 0.05 % Coomassie brilliant blue G-250. Gels were destained in 7 % acetic acid and 5 % methanol.

2.2.3.2. Nondenaturing Polyacrylamide Slab gel Electrophoresis

Nondenaturing polyacrylamide slab gel electrophoresis was performed by a modification of that described by Laemmli (292). Slab gels contained a resolving gel (13 cm) composed of 18 % acrylamide, 0.36 % N,N'-methylenebis(acrylamide), and 375 mM Tris-HC1 (pH 8.8) and a stacking gel (1 cm) containing 3 % acrylamide, 0.08 % N,N'-methylenebis(acrylamide), and 125 mM Tris-UC1 (pH 6.8). Both the resolving and stacking gels were polymerized as described above. Protein samples (30-60.tl) were adjusted to final concentrations of 50 mM Tris-HC1 (pH 6.8), 10 % (w/v) glycerol, and 0.01 % bromophenol blue prior to being loaded on the gel. Electrophoresis was performed at 4°C and 100 V until the tracking dye migrated through the stacking gel, at which point the electrical potential was increased to 200 V. Electrophoresis continued until the tracking dye migrated within -2 cm from the bottom of the gel. The native running buffer contained 25 mM Trizma base and 192 mM glycine. Protein bands were detected by a rapid protein staining method originally described by Reisner (290) with modifications as described above. 79

2.2.3.3. Urea-polyacrylamide Sequencing Gel Electrophoresis

Analysis of uracil-DNA glycosylase reaction products and whole cell extract base excision repair reaction products was performed using denaturing polyacrylamide gels (30 x 40 x 0.08 cm) composed of 12 % acrylamide, 0.40 % N,N'- methylenebis(acrylamide), 8.3 M urea, and TBE buffer (90 mM Tris, 90 mM boric acid, 2 mM EDTA). Polymerization was catalyzed by adjustment to 0.067 % (w/v) ammonium persulfate and 0.0 12 % (vlv) TEMED. Samples were mixed with an equal volume of denaturing formamide dye buffer (95 % deionized formamide, 10 mM EDTA, 0.1 % bromophenol blue, 0.1 % xylene cyanol) and heated at 95°C for 3 mm. Samples (5-15 jil) were loaded and electrophoresis was performed at 1000-1200 V in TBE buffer until the bromophenol blue tracking dye had migrated -P20-25 cm. The gels were soaked in tepid water for -2-3 mm to diffuse urea out of the gel and gels were dried under vacuum and autoradiography was performed with X-OMAT AR5 film (Kodak). Bands containing 32P radioactivity were either excised from the gel for solubilization and measurement of 32P radioactivity by liquid scintillation counting or the intact gel was quantified using a Phosphorlmager. Nondenaturing 5 % polyacrylamide sequencing gels were similarly prepared but without urea. Samples (-5-10 j.tl) were adjusted to 50 mM Tris-HC1 (pH 6.8), 10 % (w/v) glycerol, and 0.01 % bromophenol blue and electrophoresis was performed at 750 V as described above. Gels were dried under vacuum, autoradiography was performed, and 32P radioactivity was quantified using a Phosphorlmager.

2.2.3.4. Agarose Gel Electrophoresis

Agarose slab gels (0.8 %) were prepared by heating electrophoresis grade agarose powder in TAE buffer (40 mM Iris-acetate and 1 mM EDTA (pH 8.0)) in a microwave for -'1-2 mm. After cooling to --50-60°C, ethidium bromide (500 j.tg/ml) was added to 0.1 j.tg/ml, the mixture was poured into 12 x 13 cm or 7 x 8 cm casting trays (Owl Scientific) containing a 20-well or 10-well Teflon comb, respectively, and allowed to polymerize for -1 h. To physically separate Form I from Form II DNAor to measure M13mp2 DNA recoveries or primer extension reaction products, samples were combined with agarose dye buffer to a fmal concentration of 10 mM EDTA (pH 8.0), 0.1% SDS, 5% (w/v) glycerol, and 0.01% bromophenol blue, loaded onto 0.8% agarose gels, and electrophoresis was performed at 100 V (12 x 13 cm gel) or 70 V (7 x 8 cm gel) in TAE buffer containing 0.1 .tg/ml EtBr until the tracking dye had migrated -75% of the distance of the gel. DNA was visualized by UV-light transillumination (302 nm).

2.2.4. Purification ofEscherichja coliDouble-stranded Uracil-DNA Glycosylase

E.coli JM1O9 cells transformed with pKK-Dug were grown at 37 °C in 12 L of YT medium supplemented with 0.01% ampicillin. After reachinga cell density of

6.5 xi08cell/mi, 100 ml of 100 mM IPTG was added to induce dug gene expression and incubation was continued for an additional 4 h at 37 °C. Cells were harvested by centrifugation in a GSA rotor at 6,000 rpm for 15 mm at 4 °C. Cell pellets (100 g) were thawed on ice and resuspended in 300 ml of sonification buffer composed of 50 mM Tris-HC1 (pH 8.0), 1 mM EDTA, and 0.1 mM DTT. The cells were then disrupted by sonification using a sonic Dismembrator (Fisher, model 300) and the cell lysate was collected on ice and centrifuged in a SS34 rotor at 13,000 rpm for 20 miii at 4 °C to pellet the cellular debris. Ugi (0.24 ml, 3 x 106 units) was added to the pooled cell lysate (120 ml) to inactivate endogenousE.coli Ung activity. The volume of the supernatant was measured and an equal volume (380 ml) of 1.6 % streptomycin sulfate in the sonification buffer was slowly added by burette with stirring over a period of 30 mm. After equilibration for an additional 30 mm, precipitated material was removed by centrifugation in a SA600 rotor at 12,000 rpm for 15 miii at 4 °C, and the supematant was recovered and designated fraction I. Ammonium sulfate powder :J1

was slowly added to this supematant (-380 ml) to 40 % saturation and equilibrated for 10 mm. The precipitate was removed by centrifugation in a SA600 rotor at 12,000 rpm for 15 mm at 4 °C, and powdered ammonium sulfate was slowly added to the collected supernatant to 80 % saturation and equilibrated for 10 mm. The precipitate was recovered by centrifugation as described above and resuspended in -20 ml of UEB buffer (10 mM Hepes-KOH (pH 7.4), 10 mM 2-mercaptoethanol, 1 mM EDTA, 1 M NaC1 and 5% (w/v) glycerol), dialyzed overnight against the same buffer, and designated fraction II. The dialyzed sample was loaded onto a Sephadex G-75 column (6cm2x 88 cm) equilibrated in T.JEB buffer and fractions (6.48 ml) were eluted with equilibration buffer at a flow rate of 20 nil/h and assayed for Dug activity. Uracil- and ethenocytosine-DNA glycosylase activity were assayed using [32P]C'G-34-mer. Active fractions were pooled (-50 ml) and dialyzed against HAB buffer (10 mM potassium phosphate (pH 7.4), 1 mM DTT, 200 mM KC1) overnight. The dialyzed pooi was fraction III. The fraction III was loaded onto a hydroxyapatite column (19.6 cm2x 3.5 cm) equilibrated in FLAB buffer, fractions (6.48 ml) were eluted with equilibration buffer at a flow rate of 30-40 mI/h and assayed for Dug activity. Active fractions were pooled (-80 ml) and dialyzed against DAB (30 mM Tris-HC1 (pH 7.4), 1 mlvi EDTA, 1 mM DTT, and 5% (w/v) glycerol) buffer; fraction IV. The dialyzed sample was applied to a single-stranded DNA-agarose column (4.9cm2x 16 cm) equilibrated in DAB buffer, washed with 150 ml of equilibration buffer, and eluted with a 350 ml of linear gradient of 0 to 600 mM NaC1 in DAB buffer at a flow rate of

20-30 mi/h. Fractions (5 ml) were collected, and 5 .tlof aliquots from every third fraction were assayed for Dug activity and monitored for conductivity. Active fractions were pooled (-80 ml) and dialyzed against DEAF buffer (30 mM Tris-HC1 (pH 7.5), 1 mM EDTA, 1 mM DTT, and 5% (w/v) glycerol), and designated fraction V. Fraction V (10ml)was adjusted to 50 mM NaCl and loaded on the DEAF- Sephadex A50 column (4.9cm2x 6 cm) equilibrated with DEAF buffer containing 50 mM NaCl. The column was washed with the same buffer (60ml)and eluted with a 100 ml of linear gradient of 150 to 500 mM NaCl in DEAF buffer at a flow rate of 20-25 mi/h. Fractions (3 ml) were collected, and the active fractions detected in the flow through were pooled (2l ml), concentrated (-2 fold), dialyzed against DAB buffer, and the resulting preparation was designated Dug (fraction VI). The purity of each fraction (I-VT) was analyzed in 12.5 % SDS-polyacrylamide gel electrophoresis and shown in Figure 7.

2.2.5. Western Blot Analysis

Following electrophoresis, proteins were transferred to a polyvinylidene fluoride (PVDF) membrane or Immobilon-P membrane (Millipore) from the 12.5% SDS-polyacrylamide gel for Western blot analysis using a semi-dry transfer method. Six sheets of Whatman (3MM CHIR) chromatography filter paper and a membrane were cut to the dimensions of the resolving gel. One MilliBlot-Graphite Electroblotter graphite plate (anode electrode plate) was wet with anode buffer 1(0.3 M Tris, 10% methanol, pH 10.4) and the transfer apparatus was assembled by positioning on the anode plate, 2 sheets of filter paper presoaked in anode buffer I followed by 1 sheet of filter paper presoaked in anode buffer 11(25 mM Tris, 100% methanol, pH 10.4), the membrane presoaked in 100% methanol and distilled water, the resolving gel presoaked in cathode buffer (25 mM Tris, 40 mM glycine, 10% methanol, pH 9.4), and 3 sheets of presoaked filter paper in cathode buffer. A test tube was rolled across the top of the filter paper to remove air bubbles and the cathode plate was set in position to complete the apparatus. The gel was transferred at 2.5mA/cm2for 20 miii. All subsequent incubations were performed by shaking at room temperature. The membrane sheet was blocked overnight in 20-30 ml TTBS (100 mM Tris (pH 7.5), 150 mM NaC1, 0.5% Tween-20, and 5% non-fat milk). Anti-Dug serum or affinity-purified antibody was diluted (1 :5,000 and 1:10,000, respectively) in 10 ml of TTBS and incubated with the membrane for 1 h. The membrane was washed three times by incubating with 20 ml of TTBS for 5 mm. Goat anti-rabbit IgG secondary antibody conjugated with alkaline phosphatase (Boehringer Mamiheim) was diluted 1:2000 in 10 ml of TTBS and incubated with the membrane for 1 h. The membrane was washed as described and then reacted with 10 ml of 100 mMNaHCO3(jH 9.8),

10 mM MgC12 containing 100jtleach of 30 mg/mi nitroblue tetrazolium in 70% DMF and 15 mg/mi 5-bromo-4-chloro-3-indolyl phosphate in 100% DMF for 5-20 mm to develop a purple precipitate at sites of the alkaline phosphatase reaction. The reaction was terminated by washing the membrane with distilled H20. The membrane was air- dried and protected from light.

2.2.6. Purification of Polyclonal Antibody for Escherichia coli Double- Stranded Uracil-DNA Glycosylase

2.2.6.1. Polyclonal antiserum production

After a 5-day quarantine, pre-immune serum was collected by performing test bleeds (3-5 ml) via ear bleeds on female New Zealand White rabbits (2.1-2.5 kg). The blood was clotted by incubating for 1 h at 37°C, followed by 8-12 h at 4°C. The sample was centrifuged in an IEC clinical centrifuge (setting #4) at room temperature for 20 mm and the serum (2-2.5 ml) removed from the clot. Serum was stored at -20°C in 0.5 ml aliquots. By using a Virtis "23" mixer, complete Sigma Titermax classic adjuvant was emulsified with an equal volume of purified E. co/i Dug (0.3 mg in 0.5 ml ddH2O). Approximately 0.1 ml of antigenladjuvant was injected subcutaneously into each rabbit at five separate sites. Boost injections of Dug plus complete adjuvant were performed every 4 weeks as described above. Test bleeds were taken approximately 10 days after each boost and serum was processed as described. After 3 boosts a terminal bleed was performed by cardiac puncture, yielding approximately 150 ml of blood. After clotting, the samples were centrifuged at 5,000 x g for 10 mm at 4°C to recover the serum (75 ml). The affinity of anti-Dug antibody for Dug protein was estimated by Western blot analysis using the purified Dug (fraction VI) as an antigen (Figure 8). 2.2.6.2. Dug-sepharose Chromatography

Dug-Sepharose affinity chromatography was performed by incubating samples

(100 tl)with an equal volume of the resin slurry in DAB buffer (30 mM Tris-HC1 (pH 7.4), 1 mM EDTA, 1 mM DTT, 5 % (w/v) glycerol) for 10 mm at 25°C. Micro- purification columns were designed by plugging gel-loading micropipet tips with Sigma-coat-treated glass wool. The plugged tips were incubated 10 mm at 25°C with

100!.tlof 10 mM potassium phosphate buffer containing 100 pg/m1 BSA to block nonspecific protein binding. The buffer was removed by brief centrifugation in an IEC clinical centrifuge (setting #4) at room temperature and the resin-enzyme slurry was transferred to the micro-columns. The buffer was removed from the resin by centrifi.igation as described. The resin was washed three times with 200 p.1 of DAB buffer applied in two equal aliquots, while centrifuging between buffer additions. The columns were sealed by heating and 100 p.1 of SDS gel loading dye (50 mM Tris-HC1 (pH 6.8), 1% SDS, 143 mM f3-mercaptoethanol, 10% (wlv) glycerol, and 0.04% bromophenol blue) was added to the resin. The columns were incubated at 100°C for 10 nun to remove bound protein. The columns were then unsealed and centrifuged to dryness. Aliquots (25 p.1) of flow through and wash fractions and 50 p.1 of each SDS elution fraction were analyzed by 12.5% SDS polyacrylamide gel electrophoresis.

2.2.6.3. Polyclonal Antibody Purflcation

Conditions for the affinity purification of anti-Ung polyclonal IgG were optimized by suspending Dug-Sepharose resin in 10 mM phosphate buffer (pH 7.4). Polyclonal antiserum from a terminal bleed (50 p.1) was mixed with 100 p.! of Dug- Sepharose and incubated for 10 miii at 25°C. The mixture was poured into a Bio-Rad spin column and washed with 10 mM phosphate buffer (pH 7.4). Bound antibody was eluted as described in Table 5. The eluted antibody was assayed by SDS-PAGE. Aliquots (1.25 tl)were analyzed by 12.5% SDS-polyacrylamide gel electrophoresis followed by silver staining (Figure 9). Once optimal purification conditions were determined, a large scale purification was performed by mixing 12.5 ml of Dug-Sepharose with 25 ml of serum. The slurry was mixed on a rocking table for 30 mm at 25°C then transferred to a 1.5 cm diameter BioRad column. Fractions (1 ml) were collected at a rate of 15 mi/h at 4°C. The resin was washed with approximately 25 ml of 10 mM potassium phosphate buffer (pH 7.4) and bound antibody eluted with 3.5 M MgC12 in 10 mM potassium phosphate buffer (pH 7.4). Fractions were assayed for total protein by absorbance

spectroscopy at 280 nm. Aliquots of a 1:50 dilution of flow through fractions (6.25j.tl) and wash and elution fractions (1.25 p1) were analyzed by 12.5 % SDS- polyacrylamide gel electrophoresis and silver staining. Fractions containing the peak anti-Dug IgG were pooled and dialyzed against potassium phosphate buffer (pH 7.4) containing 100 mg/mi BSA at 4 °C. The pool fraction was aliquoted and stored at- 80°C. Figure 7. SDS-polyacrylamicle gel analysis of recombinant Dug isolated at various steps during the purification. Protein samples from the purification of recombinant Dug (fraction 1-VI) containing 4, 4, 3, ,3, 2, and 2 tg of protein, respectively, were analyzed by 12.5 % SDS-polyacrylamide gel electrophoresis and stained with Coomassie Brilliant Blue G-250 as described under "Experimental Procedures". The location of the protein molecular weight standards (lane lvi) for

phosphorylase b (Mr97,400), BSA(Mr66,200), ovalbumin (Mr 45,000), carbonic

anhydrase(Mr31,000), trypsin inhibitor (Mr21,500), and lysozyme (Mr 14,400) are indicated by arrows from top to bottom, respectively. The location of Dug is also indicated by an arrow. MWxIO-3 M I II lU IV V

97.4I'.- 66.2-'- 45-,-

31

21.5-'-

14.4-'- TD-'-

Figure 7 Table 5. Buffer Conditions for Dug-Sepharose Affinity Chromatography

WashBuffer" ElutionBuffet NeutralizingBuffrr

10 mM potassium phosphate 10 mM potassium phosphate 100 mM glycme (pH 2.5) (pH6.8) (p118.0) 10 mM potassium phosphate 100 mM ethanolamine 10mM potassium phosphate (pH 8.0) (pH 11.5) (pH6.8) 10 mM potassium phosphate 3.5MMgCL2,lOniM None (pH 7.4) potassium phosphate (pH 7.4) a. After incubation with sample the affinity resin was washed 3 times with 200 j.il of wash buffer applied in 100 p.1 aliquots. b. Bound protein was eluted from the affinity resin with 100 p.1 of elution buffer. c. When necessary, the eluted fraction was collected into 10 p.! of neutralization buffer. Figure 8. Western Blot analysis of affinity and specificity of anti-Dug antiserum and purified anti-Dug antibody for Dug protein. Western blot analysis was performed as described in Section 2.2.5. using various amounts of Dug protein (5, 0.5, 0.05, 0.005, and 0.001 j.tg, lanes 1-5, respectively, or 500, 50, 10, 5, 1, 0.5, and 0.1 ng, lanes 6-12, respectively) and anti-Dug antiserum from a primary immunization (1:2000 dilution, left panel) or purified anti-Dug antibody after second boost immunization and Dug-sepharose affinity chromatography (1:16000 dilution, right panel) as a probe. The location of the protein molecular weight standards (lane lvi) for

BSA (Mr 83,000 Da), carbonic anhydrase (Mr43,600 Da), soybean trypsin inhibitor

(Mr 31,700 Da), lysozyme (Mr 17,900 Da), and aprotinin (Mr7,200 Da) are indicated by arrows on the left from top to bottom, respectively. The location of Dug protein is also indicated by an arrow. Lanes

MWxIO3M 12 3 4 5 6 7 8 9 10 11 12

83 42.6- 31 7 Dug-'- - -- u-Dug 17.9-'-

7.2-'-

Figure 8 91

Figure 9. Purification of anti-Dug antibody by Dug-sepharose affinity chromatography. The purification of anti-Dug antibody by Dug-sepharose affmity chromatography was described in Section 2.2.6.3. A sample (1.25 p1) of each wash fraction (lanes 1-5) and elution fraction, by glycine (pH 2.5) (lanes 6-14), ethanolamine (pH 11.5) (lanes 15-23), andMgC12(pH 7.4) (lanes 24-32), were analyzed by 12.5% SDS-polyacrylamide gel electrophoresis, and stained with silver stain. The location of the protein molecular weight standards (lane M) for

phosphorylase b (Mr 97,400 Da), BSA (Mr66,200 Da), ovalbumin (Mr45,000 Da),

carbonic anhydrase (Mr 31,000 Da), and trypsin inhibitor (Mr 21,500 Da) are indicated by arrows from top to bottom, respectively. The location of the anti-Dug antibody, which contains heavy (top arrow) and light (bottom arrow) chains, is indicated by two arrows. Lanes MWx1O' M12345678 974-ø 66.2 :1 ...... 45 Anti Dug antibody 31

21.5

TD-t

Figure 9 93

2.2.7. Construction of Human IJNG* and UNG Overexpression Plasmids

The nucleotide sequence corresponding to the core catalytic domain of human uracil-DNA glycosylase (UNG* or UNGA84 in the literature (100)) was amplified in a polymerase chain reaction (100 j.tl) containing pUNG15 as template (0.5 j.g), primers (1 j.tM each) RI 5'-GCGAATTCTTTGGAG AGAGCTGGAAG-3' and H3 5'- GCAAGCTTTCACAGCTCCTTCCAGTC-3' as forward and reverse primers, respectively, 1 xThermoPol (New England Biolabs) buffer, 200 pM each dATP, dTTP, dCTP, and dGTP, and 2 units of Deep Vent DNA polymerase. Following digestion with EcoRI and Hindill, the UNG* fragment (700-bp) was agarose gel-purified and ligated to pTrc99A DNA that was similarly digested and purified.E. coli CYlOrec/pRP was then transformed with the ligation mixture and plated on medium containinglOO jig/mi ampicillin and 25 jig/mi chloramphenicol. Similarly, the His- tagged IJNG*, designated as UNG, DNA fragment was generated by PCR using UNG*/pTrc99A as a template, the forward primer 5'-GCGAATTCCATC ACCATCACCATCACTTTGGAGAGAGCTGGAAG-3', and the H3 reverse primer described above. The UNG gene construct was ligated into pTrc99A as described for

UNG*, and subsequently subcloned into pET-22b atNdeIandHindillsites. DNA sequence analysis verified that the nucleotide sequence of the UNG constructs was identical to that reported by Slupphauget al. (100).

2.2.8. Purification of Core Catalytic Domain of Human Uracil-DNA Glycosylase

E. coliCY1 1/pRPIUNG*pTrc99A was grown at 37°C with shaking (250 rpm) in 1.5 L of 2xYT medium containing 1 % glucose and 100 pg/ml ampicillin. Upon reaching mid-log phase growth(0D600 =-'0.6), the culture temperature was reduced to 30 °C and IPTG was added to 1 mM fmal concentration. Following incubation for 5 h at 30 °C, the cells were harvested by centrifugation and resuspended in 30 ml of 94

buffer SB (50 mM Tris-acetate (pH 7.0), 10 mM NaC1, 0.5 mM PMSF, 1 mM DTT and 10 % (wlv) glycerol), and adjusted to 500 jig/mI lysozyme (Sigma) prior to storage at -80 °C. The cell suspension was thawed and sonification conducted as described above. All operations were carried out at 4 °C unless otherwise indicated. After centrifugation at 20,000 x g for 15 mm, the supematant fraction was reserved, and the cell pellet resuspended in 30 ml of buffer SB. Both the sonification and centrifugation steps were repeated and the supematant fractions were combined to produce the cell-free extract (Fraction I). An equal volume of 1.6 % streptomycin sulfate was slowly added to the crude extract while mixing on ice for 1 h. Precipitates were then removed by centrifugation at 15,000 x g for 15 mm and the supematant fraction (-90 ml) was designated Fraction II. Fraction II was loaded onto a DEAE- cellulose column (4.9cm2x 20 cm) equilibrated in buffer SB and the column was washed with 150 ml of the equilibration buffer. Fractions containing uracil-DNA glycosylase activity were pooled (Fraction III) and applied to a CM-Sephadex column (4.9cm2x 10 cm) equilibrated in buffer SB. After washing the column with 500 ml of equilibration buffer, a 150 ml linear gradient of 10-400 mM NaC1 in buffer SB was applied. The peak of UNG* activity eluted at -150 mM NaC1; active fractions were pooled, dialyzed against buffer SB, and designated Fraction IV. Fraction IV was loaded onto a phosphocellulose column (4.9cm2x 10 cm) equilibrated in the same buffer. The column was washed with 100 ml of buffer SB and eluted with a 150 ml linear gradient of 10-400 mM NaCI in buffer SB. The enzyme activity peak that eluted at 250 mM NaCl was pooled and dialyzed against buffer SB (Fraction V). Fraction V was loaded onto a single-stranded DNA agarose column (19.6cm2x 5cm) equilibrated in buffer SB. The column was washed with 500 ml of equilibration buffer, and a 500 ml linear gradient of 10-800 mM NaC1 in buffer SB was applied. Eluted fractions were assayed for UNG* activity and a single peak of activity was detected at 175 m14 NaCl. The enzyme was pooled as Fraction VI, dialyzed against buffer SB containing 40 mM NaCl, and applied to a Poly(U) column (1.8cm2x 3 cm) equilibrated in the same buffer. After washing the column with 50 ml of equilibration buffer, the enzyme was eluted using a 50 ml linear gradient of 40-600 mM NaC1 in buffer SB. Fractions containing IJNG* activity were analyzed by 12.5 % SDS- polyacrylamide electrophoresis and pooled, constituting fraction VII (peak activity eluted at 275 mM NaC1). After dialyzing fraction VII against buffer KDP (10 mM potassium phosphate (pH 7.4), 1 mlvi DTT, 0.5 mlvi PMSF) containing 50 mlvi KC1, the sample was loaded onto a hydroxyapatite column (1.6cm2x 2 cm) equilibrated in the same buffer. The column was washed with 50 ml of equilibration buffer and proteins were eluted using a 30 ml of linear gradient of KC1 (50-600 mlvi) in buffer KDP. Uracil-DNA glycosylase activity was eluted as a single symmetrical peak (-450 mM KC1), pooled, and stored at 80 °C as Fraction VIII. The yield of the purification after each chromatographic step was shown in Table 6. The purity of each fraction (I VIII) was analyzed in 12.5 % SDS-polyacrylamide gel electrophoresis and shown in Figure 10. The molecular weight was estimated by G-75 size-exclusion column chromatography using phosphorylase b, serum albumin, ovalbumin, carbonic anhydrase, trypsin inhibitor, and lysozyme as protein standards (Figure 11).

2.2.9. Random and Site-directed Mutagenesis of Core Catalytic Domain of Human Uracil-DNA Glycosylase

Codon-specific PCR-based random mutagenesis of the UNG Arg276 (AGA) codon was carried out following the QuikChange procedure (Stratagene). Primers, 5'- CTTTGTCAGTGTATNNNGGGTTCTTTGGATG-3' (FP-3 1 -mer) and 5' CATCCAAAGAACCCNNNATACACTGACAAAGG-3' (RP-32-mer), were synthesized that contained an equal mixture of the four phosphoramidite nucleoside monomers (NNN) at the AGA codon corresponding to Arg276. PCR reaction mixtures

(50 tl)contained 50 ng of UNG/pET-22b DNA template, 14 pmol each of the forward (FP-31-mer) and reverse (RP-32-mer) primers, 250 .iM each of dATP, dTTP, dCTP, and dGTP, 2.5 units ofPfuTurbo DNA polymerase, and lxPfureaction buffer (Stratagene). PCR was carried out in a Hybaid PCR Express thermocycler according to the following parameters: 1mmdenaturation at 94 °C, 18 cycles of 94 °C (30 s), 55°C (1 mm), 68 °C (12 mm), and 72 °C (12 miii). The PCR-DNA product was then digested withDpnI(25 units) for 2 h at 37 °C. E. coli CYlOrec/pRP Z-competent cells (1-2 x iO'° cells) were thawed on ice, gently mixed with 5 j.ii of DpnI-digested DNA, incubated on ice for 1 h, spread on pre-warmed 2x YT plates containing 1 % glucose, 100 jig/mi ampicillin, and 34 jig/mI chioramphenicol, and incubated overnight at 37 °C. Site-specific PCR-based mutagenesis of the Arg276 codon of UNG was carried out for five amino acid substitutions as described above except that the forward primers (31-mers) were 5'- CTTTGTCAGTGTATXXXGGGTTCTTTGGATG-3' where XXX was TGC for cysteine, CAC for histidune, ATG for methionine, GTA for valine, and TGG for tryptophan amino acid substitutions. The corresponding reverse primers (31-mers) were 5'-CATCCAAAGAACC CYYYATACACTGACAAAG-3' where YYY was GCA, GTG, CAT, TAC, and CCA for the cysteine, histidune, methionine, valine, and tryptophan amino acid substitutions, respectively.

2.2.10. Isolation of R276 Mutants

Overnight cultures of ampicillun-resistant E. coIl CY1 Orec/pRP/UNG-pET-22b were used to inoculate (1/25 dilution) flasks containing Terrific Broth (30 ml) supplemented with 100 jig/mi ampicillin, and incubation was continued for 3 h at 37 °C. Cells were harvested by centrifugation, resuspended in 1 ml of ice-cold sonification buffer (50 mM Tris-HC1 (pH 8.0), 1 mM EDTA, 1 mM DTT, 1 mM PMSF), placed on ice and subjected to a 2 mm sonic pulse (Branson Sonifier 450, microtip, 30 % duty cycle, power setting 3). The cell lysates were clarified by centrifugation at 14,000 x g for 10 mill at 4 °C, and the protein concentration of the supernatant fraction was determined by the Bio-Rad Protein assay using bovine serum albumin as the standard. Extracts were diluted 1/10 and 1/100 in UDB buffer (50 mM Hepes-KOH (pH 7.9), 1 mM EDTA, 1 mlvi DTT, and 50 jig/mi acetylated BSA) and uracil-DNA glycosylase activity was measured as described below. Growth of cultures whose cell-extract exhibited lower than average uracil-DNA glycosylase specific activity (<10 units/mg protein) was continued, and the UNG/pET-22b plasmids isolated using a QIAGEN miniprep kit for DNA sequence analysis.

2.2.11. Purification of His6-tagged Core Catalytic Domain of Human Uracil-DNA Glycosylase and R276 Mutant Proteins

E. coil CYlOrec/pRP/UNG-pET-22b was grown in 1.0 L of 2X YT medium containing 1 % glucose, 100 jtg/ml ampicillun and 34 mg/mi chioramphenicol at 37°C with shaking. When the0D600of the culture reached 0.5, the temperature was reduced to 25 °C and incubation was continued for 20 miii before the addition of IPTG to 1 mM. After incubation for 6 h, the cells were harvested by centrifugation, resuspended in 25 ml of buffer A (50 mM sodium phosphate, pH 8.0, 500 mM NaC1, 1 mM f3- mercaptoethanol, 10 mM imidazole, 5 % (w/v) glycerol) and lysozyme was added to 500 ig/ml prior to storage at -80 °C. The cell suspension was thawed on ice, supplemented with PMSF to 1 mM, and subjected to sonification as described above. The cell lysate was similarly processed except using buffer A. Following the sonification, the cell lysate (Fraction I) was diluted to a protein concentration of 1.5 mg/mi with buffer A and applied to a Ni-NTA agarose colunm (0.8cm2x 2.5 cm) equilibrated in buffer A. The column was washed with 20 ml of equilibration buffer and step ehuted with buffer A containing 300 mM imidazole. Fractions containing IJNG activity were pooled and dialyzed against buffer B (10 mM sodium phosphate (pH 8.0), 20 mM NaCl, 1 mM DTT, 5 % (w/v) glycerol) constituting Fraction II. Fraction II was applied to a hydroxyapatite column (1.8cm2x 1.1 cm) equilibrated in buffer B. The column was washed with 20 ml of buffer B containing 200 mM NaC1 and step eluted with 20 ml of buffer B containing 300 mM NaC1. Fractions containing uracil-DNA glycosylase activity were pooled, dialyzed against buffer C (30 mM Tris- HC1 (pH 7.4), 1 mM EDTA, 1 mM DTT, 5 % (w/v) glycerol) and designated Fraction III. Fraction III was applied to a Poly(U) column (1.8cm2x 1.2 cm) equilibrated in buffer C. After washing the column with 4 ml of buffer C and then with 20 ml of buffer C containing 150 mM NaCl, the enzyme was eluted with 20 ml of buffer C containing 500 mM NaC1. Active fractions were pooled and dialyzed against buffer C (Fraction IV), and stored at -80 °C. The Arg276 mutant protein preparations were purified using the same procedure.

2.2.12. Enzyme Assays

2.2.12.1. Uracil-DNA Glycosylase Activity Assay

Standard uracil-DNA glycosylase assays (100 .tl)contained 70 mM Hepes-KOH (pH 7.9), 1 mM EDTA, 1 mM DTT, and 8.2 nmol of activated calf-thymus [uracil- 3H]DNA (195 cpm!pmol of uracil). Exogenous E. coli Ung, human UNG*,or UNG were diluted in enzyme dilution buffer (50 mM Hepes-KOH (pH 7.9), 1 mM EDTA, 1 mM DTT, 100 J2g/ml bovine serum albumin), and were introduced as samples (25 p.1). Reactions were incubated at 37°C for 30 mm and terminated on ice with 250 p.1 of 10 mM ammonium formate buffer (pH 4.2). Free [3H}uracil was separated from non- hydrolyzed [uracil-3H]DNA by applying 300 p.1 of the terminated reaction to a Dowex

1-X8 ion exchange column (0.2cm2x 2.0 cm) equilibrated in 10 mM ammonium formate buffer (pH 4.2). Columns were washed with 1.7 ml of equilibration buffer and two 1 ml fractions were collected. Fractions were combined with 5 ml of Formula 989 Fluor, mixed by inversion, and 3H radioactivity was measured by Beckman Coulter LS6500 multi-purpose scintillation counter. One unit of uracil-DNA glycosylase is defined as the amount that releases 1 nmol of uracil/h under standard conditions. Table 6. Purification of catalytic domain of human uracil-DNA glycosylase

Total Total Specific activityPurffication Yield Fraction . activity protein (units/mgx 1O) (x-fold) (%) (units/tO3) (mg) Crude Cell Extract 1190 1094 0.919 1.0 100 (1) DE52 (11) 80 674 8.43 9.2 61.6

CM52 (111) 10.4 657 63.8 69.4 60.1 Phosphocellulose P11 (IV) 5.7 599 105 114.3 54.8 ssDNA agarose (V) 1.1 509 462 502.7 46.5

Poly(U) (VI) 0.24 145 606 659.4 13.3

Ilydroxyapatite (Vii) 0.12 139 1155 1256.8 12.7 100

Figure 10. SDS-polyacrylamide gel analysis of recombinant UNG* isolated at various steps during the purification. Protein samples from the purification of recombinant UNG* (fraction 1-Vill) containing 4,4, 4,4, 3, ,3, 2, and 2 p.g of protein, respectively, were analyzed by 12.5 % SDS-polyacrylamide gel electrophoresis and stained with Coomassie Brilliant Blue G-250 as described under "Experimental Procedures". The location of the protein molecular weight standards (lane lvi) for

phosphorylase b (Mr 97,400), BSA(Mr66,200), ovalbumin(Mr45,000), carbonic anhydrase(Mr31,000), trypsin inhibitor(Mr21,500), and lysozyme(Mr14,400) are indicated by arrows from top to bottom, respectively. The location of UNG* is also indicated by an arrow. 101

MWxIO-3 M I H III IV V VI VII

97.4 - = - 66.2 - 45- 31 - - i_UNG* 21.5 _ 14.4 $ TD--

Figure 10 102

Figure 11. Determination of the polypeptide molecular weight of UNG* by gel filtration chromatography. A standard curve (log M. versus Rf) was generated based on the relative mobility of prestained protein molecular weight markers for

phosphorylase b (Mr 111,000 Da), BSA (Mr77,000 Da), ovalbumin (Mr48,200 Da),

carbonic anhydrase(Mr33,800 Da), trypsin inhibitor (Mr 28,600 Da), and lysozyme

(Mr 20,500 Da). The apparent molecular weight of UNG* was determined based on the Rf for the mid-point of the gel slice (horizontal line) as indicated by the vertical arrow. 103

101 Phosphory,aseb 0181 I Serum A1bumn X 61 I I. .c UN 41 OvabumIn S I Carborncanhydrase . I 0 21I Trypsin inhibitor .01I I Lysozyme

0.2 0.4 0.6 0.8

Figure 11 104

2.2.12.2. Double-strand Uracil-DNA Glycosylase Activity Assay

The ethenocytosine-DNA glycosylase activity of Dug was determined utilizing 5'-end 32P-labeled double-stranded C'G-34-mers with the 5'-end 32P-label positioned

in the substrate containing DNA strand. Standard reaction mixtures (10j.tl)contained 25 mM Hepes-KOH (pH 7.9), 0.5 mM EDTA, 1 mM DTT, 50 mM KC1, 0.01 mM ZnCl2, 0.1 mg/mi acetylated bovine serum albumin, 0.1 pmol 5'-end 32P-labeled 34- mers, and various amounts of the enzyme orE.coli cell free extract (2tl).Where it appropriate, enzyme samples or cell free extracts were diluted with DAB buffer. Reactions were incubated at 30 °C for 30mmand then subjected to treatment with eitherE.coli endonuclease IV to cleave the AP-sites generated by base removal from DNA substrates. To conduct enzymatic cleavage of AP-sites, the reactions were terminated by heating at 70 °C for 3mmand then incubated with 0.1 units of endonuclease IV (1tl)for 30 mm at 30 °C. Endonuclease IV was then inactivated by incubation at 70 °C for 3mm.Samples were combined with an equal volume of denaturing formamide dye buffer and analyzed using denaturing 12% polyacrylamide/8.3 M urea gel electrophoresis as described in Section 2.2.2.3. The gel was then imaged by a Phospholmager (Molecular Dynamics) and the amount of 32P- labeled substrate (34-mer) and product (1 5-mer) oligonucleotide bands were quantified using ImageQuant 5.0 software (Molecular Dynamics) after subtracting background intensity of the image. During the purification of recombinant Dug, the enzymatic activity was assayed for both uracil- and ethenocytosine-DNA glycosylase using [32P]U.G- and [32P]EC'G-34-mers, respectively, and the reactionwas performed as described above except (i) reaction mixtures contained 5 tlof each fraction; (ii) 0.1 pmol of [32P]34- mer was used as substrate (iii) incubation occurred at 30 °C for 30 mm. After incubation, the reaction was terminated with an equal volume of a stop solution containing 2 % SDS and 50 mM EDTA. Samples were adjusted to a final concentration of 0.3 mg/mI yeast tRNA and 2 M ammonium acetate, extracted twice 105

with an equal volume of phenol/chloroform (50:50), ethanol-precipitated at -70°C for 20 minutes, and resuspended in 20 t1 of distilledH20.DNA samples (5 tl) were treated with E. coli endonuclease N, denaturing polyacrylamide gel electrophoresis was conducted, and [32P]DNA bands were detected as described above.

2.2.12.3. Nuclease Assay

Reaction mixtures (100 j.tl) contained 8-200 ng of wild-type or mutant UNG (Fraction IV), 100 j.tg/ml acetylated BSA, buffer C, and 12.5 ng of single-stranded carboxyfluorescein (FAM) 5'-end labeled oligonucleotide: 5'-FAM-T-25-mer or 25 ng of double-stranded oligonucleotide: 5'-FAM-TA-25-mer. Following incubation at 37 °C for 30 mm, the reaction mixture was mixed with an equal volume of 2x sample buffer (95% deionized formamide. 1mM EDTA, and 0.05% bromophenol blue), and subjected to 15 % polyacrylamide, 8.3 M urea gel electrophoresis. The gels were scanned using an FMBioII fluorescence imaging system (Hitachi Genetic Systems) and the fluorescent bands were analyzed with IlniageQuant software (Amersham Biosciences).

2.2.12.4. UNGUgi Binding Assay

Ung (fraction V), UNG, or R276X (Fraction N) mutant protein (40 pmol) was combined with Ugi (100 pmol) in a reaction mixture (15 p1) containing buffer C and incubated at 20 °C for 10 mm and then at 4 °C for 20 miii. Following complex formation, 120 p1 of 40 mM CAPS-NaOH (pH 10.5) and 30 tl of buffer LB (300 mM CAPS-NaOH, pH 10.5,25% sucrose and 0.04 % bromphenol blue) were added to each sample. Non-denaturing polyacrylamide slab gel electrophoresis was performed by a modification of the procedure described in Section 2.2.3.3. Briefly, samples (100 were loaded onto a nondenaturing 10 % polyacrylamide gel, polymerized using 0.3% ammonuum persulfate, that contained 150 mM CAPS-NaOH (pH 10.5) in place 106

of Tris-HC1 (pH 8.8) buffer. Electrophoresis was conducted at 4°C and 250 V with 40 mM CAPS-NaOH (pH 10.5) running buffer until the tracking dye migrated -8 cm. Protein bands were detected after staining the gel for 1 h in dye solution containing 0.04 % Coomassie brilliant blue G-250 in 2.5 % HC1O4. Gels were destained with 5 % acetic acid and then imaged with a UVP ImageStore 7500.

2.2.13. UV-Catalyzed Photocrosslinking Reaction

Three oligonucleotides were 5'-end 32P-labeled in reaction mixtures containing 50 mM Tris-HC1 (pH 7.5), 1 mM EDTA, 5 mM DTT, 10 mM MgCl2, 330 pCi of [y- 32P]ATP (6000 Ci/mmol), 200 units of T4 polynucleotide kinase and 35 nmol of dU- 25-mer, T-25-mer (5'-GGGGCTCGTATAAGGAATTCGTACC-3'), or U-25-mer (5'- GGGGCTCGTAUAAGGAATTCGTACC-3'). After incubation for 20 mm at 37°C, ATP was added to 1 mM and incubation was continued for 30 mm. Unreacted [y- 32P]ATP and ATP were removed usinga P-4 Bio-Gel (Bio-Rad) spun column. The 32P-labeled oligonucleotideswere concentrated, buffer exchanged into TE buffer using a Centricon 3 centrifugal filter unit (Millipore), and adjusted to 20 pM in TE buffer.

Samples (12p1)containing 20 pmol of 32P-labeled oligonucleotide (18,000 cpm/pmol), 40 pmol of TJNG (Fraction IV), and buffer C, were placed on ice and TJV- irradiated (?max= 254 tim) in a Stratalinker 1800 (Stratagene) for various times as indicated in the figure legends. Following UV-irradiation, 3 pA of buffer S (45 mM Tris-base, 445 mM boric acid, 1 mM EDTA, 25 % sucrose, 0.04 % bromphenol blue) was added and samples (6 p1) were analyzed by nondenaturing 10 % polyacrylamide gel electrophoresis. The gels were dried and the amount of 32P radioactivity detected in each band was determined using a Phosphorlmager and ImageQuant software (Molecular Dynamics). 107

2.2.14. Generation of Uradil-containing Concatemeric DNA Subsfrate

Oligonucleotides, U-30-mer (5'- GCGTGACGCACTGAUAAGTGAATTCGACCG-3') and A-30-mer (5'- CGTCACGCCGGTCGAATTCACTTATCAGTG-3') were 5'-end phosphorylated as described above, except that 50 mM NaC1 was included in each reaction. [y-32P]ATP was added to the U-30-mer reaction, whereas nonradioactive ATP was included in the A-30-mer reaction. Oligonucleotides [32P]U-30-mer and A-30-mer were annealed in a Hybaid PCR Express Thermocycler using the following parameters: 2 mm each at 85, 75, 65, 60, 55, 50, 45, 40, 35, 25, 15, and 5 C. Duplex monomer (30-mer) units were joined by ligation in a reaction mixture (4.4 ml) containing buffer K (5 mM Tris-HC1 (pH 7.5), 1 mM EDTA, 5 mlvi DTT, 10 mMMgC12) supplemented with 1 mM ATP,

440 units of T4 DNA ligase, and 58 nmol of[y-32P]U/A-30-mer. The ligation reaction mixture was divided among 12 microcentrifuge tubes (367tleach) and incubated in a Hybaid thermocycler programmed to carry out 198 cycles of 1 mm at 10 °C followed by 1 mm at 22 °C. After thermocycling, incubation was continued at 16 °C for 6 h, and the reaction mixtures were pooled and adjusted to 5 % glycerol (w/v). The reaction mixture was loaded onto a Sephacryl S-500 column (1.8cm2x 70 cm) equilibrated in TE buffer containing 150 mM NaC1 and eluted with equilibration buffer. Fractions (1 ml) were collected and analyzed using non-denaturing 4 % polyacrylamide gel electrophoresis followed by autoradiography. Fractions that contained high molecular weight (>600 bp) concatenated [32P]DNA were pooled, concentrated and buffer-exchanged into TE buffer using Centricon 30 centrifugal filters (Millipore). The concentration of the [32P]DNA substrate was determined spectrophotometrically (10D260nm=50 p.gIml). 2.2.15. Restriction endonuclease Digestion of the Uracil-containing Concatemeric Polynucleotide Substrate

Restriction endonuclease reaction mixtures (10j.tl)contained concatenated [32P]UA-DNA (10 pmol),EcoRI(10 units) or HpaII (10 units), and either lx EcoRJ buffer or NEB buffer 1 (both New England Biolabs), respectively. After incubation for 60 mm at 37 °C, the reactions were terminated by the addition of 3 MK2HPO4(pH 13.7) to 0.3 M and heating at 95 °C for 30 mm. The samples were then neutralized by adding 1.5 MK112PO4to 0.3 M followed by 13.3 p.1 of TE buffer. The concatenated [32P]DNA (10 pmol)was also digested with UNG (70 units) for 60 mm at 37 °C in two reaction mixtures (10 p.1 each) containing 70 mM HEPES-KOH (pH 7.9), 1 mM EDTA and 1 mM DTT; reactions were terminated with Ugi (1 p.g). One reaction was subjected to alkaline hydrolysis followed by neutralization as described above; the other reaction was divided into two aliquots (5.5 p1). One aliquot was supplemented with TE buffer (4.5 p.1) and the other was adjusted to lx EcoRI buffer for incubation with EcoRI (10 units) for 60 mm at 37 °C. The reaction was terminated, and both samples were processed as described above. An equal volume (26.7 p.1) of denaturing sample buffer (95 % deionized formamide, 1 mM EDTA, 0.04 % bromphenol blue) was added to each sample. Portions (6 p.1) of each sample were subjected to denaturing 12 % polyacrylamide, 8.3 M urea gel electrophoresis until the tracking dye migrated 17 cm. Gels were dried and used to expose Phosphorlmage screens which were subsequently scanned with a PSI Phosphorlmager (Amersham Bioscience). The scanned images were quantified using the ImageQuant program.

2.2.16. Uradil-DNA Glycosylase Processivity Assay

The processivity of UNG or Arg276 mutants on the double-stranded uracil- containing concatemeric [32P]DNA substrate was measured in standard reaction mixtures (100 p.1) containing 80 pmol of [32P]U.A-DNA (48.6 xiO4cpmlpmol) and 109

0.06-1.0 unit of enzyme. Incubation was performed at 37 °C and samples (10 tl)were removed at various times (0-80 mm). Reactions were stopped with the addition of Ugi

and placed on ice. Each sample was divided into two (5.5 .tl)aliquots. TE buffer (4.5

tl)was added to the first aliquot and the sample was processed using the alkaline hydrolysis/neutralization treatment described above. The second aliquot was adjusted to lx EcoRiI buffer, digested with EcoRJ (10 units) for 60 mm at 37 °C, and then similarly processed. Portions (6 s.d) of each reaction were subjected to analysis by 12% polyacrylamide, 8.3 M urea gel electrophoresis and autoradiography as described above.

2.2.17. Steady-state Fluorescence Measurements

Excitation and emission spectra of the 2AP-containing oligonucleotides were measured using an SLM-Aminco model 8100 series 2 fluorescence spectrometer equipped with a quartz cell cuvette (1 cm x 1 cm). All experiments were performed in 2 ml reaction in Buffer C and maintained at 25°C with Thermomix 1441 refrigerated water bath (Precision Scientific, Chicago, IL). The excitation spectra of 2AP- containing oligonucleotides were recorded over the wavelength range of 250-350 tim with an excitation wavelength at 310 tim, while emission spectra were recorded over the wavelength range of 330-500 tim with an excitation wavelength at 310 nm. The spectral band-pass slit was 8 tim for the emission spectra. For the time-based scans of 2AP-containing oligonucleotide samples, the excitation wavelength was set to 310 tim and the emission was monitored at 370 nm with sampling at 1 s intervals for 60 s. Samples were incubated 10 mm at 25°C prior to fluorescence measurements, which were done in triplicate. 110

2.2.18. Dissociation Constants for DNA Binding

The dissociation constant (Kd) for duplex dU2AP-25-mer (100 nM) was determined by measuring 2AP fluorescence enhancement with varying amounts (0-2 M) of UNG or R276X mutant protein. The binding data were fitted to Equation 1 after subtracting the background UNG or R276X mutant protein fluorescence intensity from each measurement:

100 x (1Z) x (2Lo) F = ______(Equation 1) (Ro+Lo+KD)+.j(Ro+Lo+Kd)2 4RoxLo

In Equation 1, F represents the % fluorescence enhancement, Lo and Ro are the total concentrations of DNA substrate and IJNG or R276X mutant protein, respectively; Kd is the overall dissociation constant for UNG or R276X mutant protein binding to dU2AP-25-mer, and Z represents the ratio of UNG- or R276X-bound dijiU2AP-25- mer fluorescence to free dU2AP-25-mer fluorescence.

2.2.19. Pre-steady State fluorescence Measurement

Stopped-flow fluorescence experiments were performed using an Applied Photophysics Stopped Flow Spectrophotometer (model SX1 8MV) in the two-syringe mode (dead time = 1.0 ms). All experiments were performed in Buffer C and maintained at 25°C with a Neslab RTE-1 11 refrigerated water bath (ThermoNeslab, Newington, NIH). To measure changes in protein tryptophan fluorescence, samples were excited at 290 nm, slit width 7 nm, and the fluorescence emission was monitored using a 320-nm long pass filter (Applied Photophysics, Leatherhead, United Kingdom). For each preparation, 10 kinetic traces were recorded and averaged for kinetic analysis (see below). All reactions, except the titration of oligonucleotide 111

concentration experiment, were performed under pseudo-first-order conditions with DNA added in at a 10-fold excess over the enzyme.

2.2.20. Data Analysis

Values for the various parameters were derived by nonlinear curve fitting using the computer program Origin version 7.0 (MicroCal Software Inc., North Hampton, MA). The rate constant of enzyme conformational change was determined using a fixed amount of UNG or R276X mutant protein combined with an excess amount of DNA substrate. The averaged tryptophan fluorescence traces were fit to Equation 2:

F. = AFie1t + AF2et+ AF(1 e") + C (Equation 2)

In Equation 2, F represents the fluorescence at time t, AFi is the amplitude of the initial fluorescence decrease,k1is the observed initial rate for the tryptophan fluorescence decrease,iF2,is the amplitude of the fluorescence decrease of the second kinetic phase,k2,is observed rate for the second phase fluorescence decrease, AF3, is the amplitude of the fluorescence recovery phase, k3 is the rate constant of this pahse, C is the constant offset, and t is the reaction time. Due to the multi-phasic (> 3) nature of the typical tryptophan fluorescence trace, accurate resolution of the individual exponentials was not possible, and the best fits obtained represent estimates of the reaction rate constants. The binding kinetics of UNG or UNG2 for U'2AP-25-mer was obtained by measuring the rate of 2AP fluorescence enhancement at a fixed concentration of UNG2. The averaged kinetic traces were fit to Equation 3:

F. = AF(1 - e_klt ) + AF2e2( +C (Equation 3) 112

In equation 3, F, represents the fluorescence at time t,iFis the amplitude of the first

kinetic phase,k1is the observed initial rate for the 2AP fluorescence enhancement,

AF2is the amplitude of the second kinetic phase,k2is an observed rate for that phase fluorescence decline after enhancement, and C is the constant offset. The UNG or UNG2 concentration dependence for the observed rates of DNA binding was fit to the Michaelis-Menten equation:

kObS + C (Equation 4)

In Equation 4,k0brepresents the initial rate,kis the maximal velocity, A is the variable substrate concentration, and K is the Michaelis constant for the variable substrate, and C is the constant offset.

2.2.2 1. Uradil-DNA Glycosylase Activity Assay (Oligonucleotide Assay)

Uracil-DNA glycosylase activities of TJNG and R276X mutant proteins were determined with single-stranded 5'-FAM-U-25-mer or duplex 5'-FAM-UA-25-mer.

Standard reaction mixtures (20jtl)contained 25 mM Hepes-KOH (pH 7.9), 0.5 mM EDTA, 1 mM DTT, 1.6 pmol- 4.4 fmol of wild-type UNG or R276X mutant protein (Fraction IV), 100 p.g/ml acetylated bovine serum albumin (BSA), and 5 pmol of either single-stranded 5'-FAM-U-25-mer or double-stranded 5'-FAM-UA-25-mer. Following incubation at 37 °C for 15 mm, enzyme reactions were stopped with the addition of Ugi (1 p.g) and placed on ice. Reaction mixtures contained single-stranded

5'-FAM-U-25-mer were supplemented with 2.4j.tlof buffer containing 3 M NaOH and 0.3 M EDTA, subjected to heat treatment at 90 °C for 45 mm, and neutralized by the addition of 2.4j.tlof 3 M HC1. Reaction mixtures containing duplex 5'-FAM-UA-25- mer were supplemented with 2.4j.tlof 0.5 M KC1 to yield a final KC1 concentration of 50 mM, and then subjected to E. coli endonuclease IV (1 unit) treatment at 37 °C for 113

30 mm. Enzyme reactions were terminated by heating reaction mixtures at 70 °C for 3 mm and then placed on ice for 3 mm. Samples of the reaction mixtures were combined with an equal volume of 2x denaturing formamide dye buffer containing 95% de- ionized formamide, 1 mM EDTA and 0.05% bromphenol blue, and the reaction products were resolved by 12 % polyacrylamide, 8.3 M urea, gel electrophoresis. The gels were scanned with an FMBioII fluorescence imaging system (Hitachi Genetic Systems) and the fluorescent bands quantified by ImageQuant software (Amersham Biosciences). The percentage of product formed was calculated by dividing the amount of 5'-FAM-lO-mer (product) by that of 5'-FAM-lO-mer plus 5'-FAM-25-mer (substrate) and multiplying by 100. Enzyme preparations were diluted in 25 mM Hepes-KOH (pH 7.9), 0.5 jtM EDTA, 1 mM DTT, 100 p.g/ml acetylated BSA.

2.2.22. Purification of Human Uracil-DNA Glycosylase, Nuclear Form

Human uracil-DNA glycosylase, nuclear form (UNG2), containing a cleavable histidine tag on the N-terminus was overproduced in BLRIpRIL/UNG2-pET28a and purified as following. Briefly, BLR/pRIL/UNG2-pET28a was grown in 2xYT medium containing 1 % glucose, 100 j.tg/ml ampicillin and 34 jtg/ml chioramphenicol at 37°C with shaking (250 rpm). When the 0D600 of the culture reached 0.5, the temperature was dropped to 26 °C and incubation was continued for 20 mm before the addition of IPTG to 1 mM. After 5 h at 26 °C cells were harvested by centrifligation, resuspended in 20 ml of buffer A (50 mM NaPO4, pH 8.0, 500 mM NaC1, 5 % glycerol (w/v), 10 mM NaHSO3, pH 8.0, 1 mM J3-mercaptoethanol, 10 mM imidazole, 2mM benzamidine) per liter of culture, and lysozyme (10 mg/mi, Sigma) was added to 100 p.g/ml prior to storage at -70 °C. The 65O ml of cell suspension was thawed, supplemented with PMSF to 1 mM, and subjected to sonification on ice. The cell lysate was clarified by centrifugation (20,000 x g), the supematant reserved, the cell pellet resuspended in buffer A (-P200 ml) and subjected again to sonification. Following centrifugation, the pellet was discarded and the supematants combined and 114

diluted to'1 .5 mg/mi protein concentration with buffer A. The diluted cell lysate (-

1.8 L), designated as fraction I, was applied (30 ml/h) to a QiagenNi2NTA agarose column (4.9cm2x 5 cm) equilibrated in buffer A at 4CC. The column was washed with 150 ml of buffer A and eluted with 150 ml of buffer A containing 300 mM imidazole. Fractions (5 ml) containing UNG2 were pooled (52 ml), dialyzed against

buffer B (10 mM NaPO4,pH 8.0, 20 mM NaC1, 1 mM DTT, 5 % glycerol (wlv), 10 mM NaHSO4, pH 8.0,2 mM benzamidine), and designatedas fraction II. After dialysis, fraction II was applied to a Q-Sepharose column (1.77cm2x 3 cm), which was directly connected to a hydroxyapatite (HA) column (1.5625cm2 x2.5 cm), equilibrated in buffer B. The Q-Sepharose columnwas washed with 20 ml of buffer B before disconnected from the HA column. The HA column was then washed with 50 mM of buffer B. Proteins were eluted with 120 ml of 0-1 M NaCl linear gradient in buffer B. Fractions (5 ml) enriched for UNG2 were pooled (-'107 ml), dialyzed against buffer C (30 mM Tris-HC1, pH 7.4, 1 mM EDTA, 1 mM DTT, 5 % glycerol (w/v), 10 mM NaHSO4, pH 8.0, 2 mM benzamidine) and designatedas fraction III. After dialysis, fraction III was applied (30 mI/h) toa Poly (U) column (0.5625cm2x 3.5 cm) equilibrated in buffer C. The column was washed (50 mi/h) first with 50 ml of buffer C, then eluted with 100 ml of 0- 1M NaC1 linear gradient in buffer C (protein l.c. peak was at -'220 mM NaC1). Fractions (-'40 ml) containing UNG2were pooled, dialyzed against buffer C, and designated as fraction IV. Fraction IVwas applied (30 mi/h) to a Macro-Prep CM column (4.9cm2x 1.5 cm) equilibrated in buffer C. The column was washed (50 mI/h) first with 40 ml of buffer C, then eluted with 75 ml of 0-1M NaCl linear gradient in buffer C (protein 1.c. peak was at -'180 mM NaCl). Fractions (-'23 ml) containing UNG2 were pooled, dialyzed against thrombin cleavage buffer (20 mM Tris-HCL, pH 8.4, 150 mM NaC1), and designatedas fraction V. Fraction V was subjected for thrombin cleavage reaction to remove the His-tagon the amino-terminus of UNG2. The thrombin cleavage reaction followed the protocol from Novagen's Thrombin Kits. Briefly, Fraction V was divided into 1 ml aliquots in 1.5 ml of microcentrifuge tubes and supplemented withCaCl2to 25 mM and 1 unit of 115

thrombin. Reaction mixtures were incubated at 20CC for 2 hour. The reactionwas then stopped by adding EDTA and benzamidine to 10 and 2 mM, respectively in the mixture. Reaction mixtures were pooled (-'27 ml) and dialyzed again buffer C. The

protein pool was applied (30 mi/h) to a Macro-Prep CM column (4.9cm2x 1.5 cm) equilibrated in buffer C. The column was washed (50 mI/h) first with 40 ml of buffer C, then eluted with 75 ml of 0-1M NaC1 linear gradient in buffer C. Fractions (-41 ml) containing purified UNG2 were pooled, dialyzed against buffer C, and designated as fraction VI. Fraction VI was divided and protein aliquots were frozen and stored at 70 ,C.

2.2.23. Matrix-assisted Laser DesorptionIlonization Mass Spectrometric Analysis

MALDI mass spectrometry was conducted using a custom-built time-of-flight mass spectrometer equipped with a two-stage delayed extraction source by the Mass Spectrometry Facilities and Service Core Unit (Environmental Health Science Center,

Oregon State University). Approximately 1j.tlof purified recombinant Dug (fraction

VI, 0.16 mg/mi) or UNG2 (fraction VI) was mixed with 3 j.tlof 4-hydroxy-a- cyanocinnamic acid in 0.1% trifluoroacetic acid, 33% acetonitrile. A droplet (--'0.5 tl) of this analyte/matrix solution was deposited on a precrystallized matrix sample probe and allowed to air-diy. Mass spectra were produced by irradiation of the sample with 30 individual laser pulses and the summed signals were calibrated using ions from an external calibrant. 116

3. Mutational Analysis of Arginine 276 in the Leucine-loop of Human Uracil-DNA Glycosylase (I)

Although the role of the conserved leucine-loop residues His268, Ser270, and Leu272 in enzyme catalysis has been examined (111,116), the function of Arg276 at the C-terminal end of the leucine-loop is still unknown. Examination of the X-ray crystallographic data (Figure 12) generated by Tamer and co-workers suggested that Arg276 might play two roles in UNG:DNA interactions: 1) The th of the Arg276 guanidinium side chain (nitrogen atoms, blue balls) may interact directly (black rippled lines) with the 5' phosphate of the cytosine residue (oxygen atom, red ball), as stated for cleaved UG DNA by Slupphaug etal. (115); and 2) The iN can participate in water-bridged (water, black ball) hydrogen bonding (dashed lines) with both the N3 of adenine (blue ball) and the carbonyl group (red ball) of Leu272 (116). This interpretation of the structural data suggested that the function of Arg276 involved interactions with the uracil-containing DNA strand on the periphery of the active site and stabilization of the Leu272 side chain either before or after it was inserted into the DNA minor groove. This chapter presents the results of mutational analysis of Arg276 in the leucine-loop of human uracil-DNA glycosylase. PCR-based random and site-directed mutagenesis was used to make a library of eighteen amino acid changes at Arg276 of the core catalytic domain of human uracil-DNA glycosylase (UNG). The mutant proteins were analyzed with respect to catalytic activity, DNA binding affmity, and ability to locate uracil residues using a processive search mechanism. The results show that mutations at Arg276 residue significantly reduced UNG catalytic activity and DNA binding affinity. In addition, UNG was demonstrated to utilize a processive search mechanism to remove uracil residues located along the same DNA strand and mutations. 117

Figure 12. Tertiary structure of human uracil-DNA glycosylase bound to DNA. Co-crystal structure of the core catalytic domain of human uracil-DNA glycosylase (LJNG*) bound to DNA containing the uncleaved uracil analog 2'-deoxypseudouridine (ijiU) (117). The DNA is shown in yellow and the view is looking into the major groove. A, Three distinct amino acid sequences (red tubes) of the UNG* polypeptide backbone (silver tubes) critical to the proposed "pinch-pull-push" catalytic mechanism (121) are shown. The 4-Pro-Ser loop (165-PPPPS-169) and the Gly-Ser loop (246-GS- 247) compress ("pinch") the deoxyribose phosphate backbone from the 5' and 3' directions, respectively (116). The Leucine-loop (268-}IPSPLSV'YR-276), which contains Arg276, penetrates the DNA base stack ("push") and occupies the helical space of the flipped-out i1jU residue (116). Conserved amino acid residues (Gln144, Asn204, and His268) in the IJNG* binding pocket capture ("pull") and stabilize the expelled extrahelicalU. a-Helices are depicted as silver cylinders and n-sheets are illustrated as blue strands. B, Ball-and-stick diagram of the UNG Leucine-loop (27 1- PLSVYR-276) shown in silver and a portion of the oligonucleotide sequence 3'- CTAU-5' shown in yellow. The eN of the Arg276 guanidinium side chain (nitrogen atoms, blue balls) is shown as interacting (black rippled lines) with the 5' phosphate of the cytosine residue (oxygen atom, red ball), as stated for cleaved UG DNA by Slupphaug et al. (115). The iN participates in water-bridged (water, black ball) hydrogen bonding (dashed lines) with the N3 of adenine (blue ball) and the carbonyl group (red ball) of Leu272 as shown in Parikh et al. (116). Structures were drawn with the Cn3D 4.0 software program using the PDB file 1EMH (MMDB: 13471) deposited by Parikh et al. (117) in the Molecular Modeling Database of the National Center for Biotechnology Information. 118

A. B.

Arg 276 G.;FfSL 4-Pro-Serb

Tyr 275 T < 44u272 Vat 274 c

Pro271

Figure 12 119 at Arg276 residue did not alter this attribute. Taken together, the results of this study establish the importance of Arg276 in mediating UNG-DNA interactions.

3.1. Results

3.1.1. Overproduction and Purification ofUNGand Arg2 76 Mutant Proteins

Oligonucleotide-directed codon-specific random mutagenesis and site-specific mutagenesis were performed on Arginine 276 of the humanUNGgene and eighteen Arg276 (R276) mutants were isolated by activity screening as described in Section 2.2.10. To facilitate the mutational analysis of human uracil-DNA glycosylase, the recombinant core catalytic domain of human uracil-DNA glycosylase (UNG*), IJNG* containing six N-terminal histidine residues (UNG), and the R276 mutant UNG proteins were overproduced in E. co/i and purified as described under Section 2.2.11. To examine the purity and relative gel mobility of each protein preparation, 4 g of IJNG*, TJNG and R276 mutant proteins (Fraction W) were subjected to 12.5% SDS- polyacrylamide gel electrophoresis. Analysis of the Coomassie-stained gel showed that UNG*, UNG, and R276 mutant proteins were purified (>95 %) to apparent homogeneity and exhibited mobility consistent with the predicted molecular weight (Figure 13A). The purified protein preparations were found to be free of contaminating nuclease activity in assays utilizing 5'-end carboxyfluorescein-labeled single- and double-stranded oligonucleotides (Figure 13B).

3.1.2. Ability of R276X Mutant Proteins to Bind Ugi

The uracil-DNA glycosylase inhibitor protein (Ugi), encoded by the B. subtilis bacteriophage PBS2, inactivates Ung-family uracil-DNA glycosylases by forming an essentially irreversible protein:protein complex with 1:1 stoichiometry (171). The X- ray crystal structures of the core catalytic domain of human uracilDNA glycosylase 120 and E. coli uracil-DNA glycosylase in complex with Ugi show that Ugi binds tightly to the sequence-conserved DNA-binding groove of the enzymes in a manner described as protein mimicry of DNA (114,183). Therefore, Ugi was used as a probe to determine whether the R276 mutant proteins were properly folded, reasoning that Ugi would only bind to enzymes with a structurally intact DNA-binding region. Each mutant protein was reacted with a small molar excess (2.5-fold) of Ugi and the mixture was resolved by nondenaturing polyacrylamide gel electrophoresis (Figure 14). Inspection of the gel shows that each R276 mutant protein analyzed formed a stable complex with Ugi, and that the appearance of the UNGUgi complex band was concurrent with the disappearance of the UNG band (Figure 14, +Ugi lanes). These results demonstrate that mutagenesis of Arg276 did not perturb the active site structure of the mutant proteins.

3.1.3. Uracil-DNA Glycosylase Activity of R2 76 Mutants

To determine the specific activity of UNG*, UNG, and R276X mutant proteins, standard uracil-DNA glycosylase reactions were conducted with 0.04-0.13 units of each protein preparation as described under Section 2.2.12.1. The specific activity (units/mg) of each preparation was then calculated and expressed as a percentage of the specific activity of UNG, which was set to 100 % (Figure 15). All eighteen R276 mutant proteins showed a reduction in specific activity. The most active mutant was histidine (R276H, 43 %), whereas the least active was glutamic acid (R276E, 0.6 %) (Figure 15, bars A-Y). Mutation of arginine 276 to the aromatic amino acids phenylalanine, tryptophan, or tyrosine resulted in differential loss of activity, as R276F, R276W, and R276Y exhibited 6, 32, and 6%, respectively, of UNG specific activity. Similarly, mutation of arginine 276 to amino acids with aliphatic side chains (R276A, R276G, R276I, R276L and R276V) resulted in 17, 4, 9, 31, and 15% of UNG specific activity, respectively. Mutation of arginine 276 to amino acids with sulfur- or hydroxyl-containing side chains (R276C, R276M, R276S, and 121

R276T) produced 21, 33, 15, and 42 % of UNG specific activity. Finally, acidic amino acid substitutions (R276D, R276E, R276N, and R276Q) and substitution with proline (R276P) resulted in 15, 0.6, 17, 12, and 11 % of TJNG specific activity, respectively (Figure 12). UNG* was more active (135 %) than UNG (100 %).

3.1.4. Effect ofR276 Mutations on 2-Aminopurine Fluorescent Intensity

Crystallographic studies by Parikh and co-workers (117) demonstrated that IJNG* was unable to cleave the Cl-C 1' bond between the pseudouracil base (iU) and its cognate 2'-deoxyribose. However, it was reported that substitution of NJU for uracil did not affect the deoxyribose pucker, aromaticity, or hydrogen-bonding interactions of the pseudo-substrate with the enzyme, and that the U base was flipped out of the double helix and into the UNG* active site pocket (117). Kinetic studies by Stivers and co-worker (118) showed an increase (2-8-fold) in 2-aminopurine (2AP) fluorescence when E. coli Ung bound uracil substrate analogs positioned adjacent or opposite to 2AP (118). Since 2AP fluorescence is quenched when the base analog is stacked within the DNA helix, the increase in 2AP fluorescence indicated that the uracil base adjacent to 2AP was flipped out (118). Based on these reports, it was reasonable to hypothesize that dU positioned opposite 2AP (dNJU.2AP-25-mer), when flipped out of the DNA helix by UNG, would also result in quantitative 2AP fluorescence enhancement. An increase in 2AP fluorescence concomitant with UNG addition was indeed observed, as is shown in the Figure 1 6A (labeled "I"). In order to determine whether the enhancement of 2AP fluorescence was dependent on UNG interactions with dijiU2AP-25-mer, an excess of uracil-DNA glycosylase inhibitor protein (Ugi) was added to the reaction to abolish UNG DNA-binding (171). Following supplementation with Ugi (Figure 1 6A, II), 2AP fluorescence was quenched (AF2) to the level (AF3) observed for Ugi alone (Figure 1 3B III). Thus, the increase in 2AP fluorescence intensity upon UNG addition (AF1) was equal to the decrease in fluorescence intensity caused by Ugi (AF2) plus the UgiIDNA fluorescence 122 background (AF3): i\F1 = AF2 + AF3. This result showed that the observed 2AP fluorescence enhancement was the result of UNG binding to dijU.2AP-25-mer. The dependence of the 2AP fluorescence enhancement on UNG concentration was determined using reaction mixtures containing 0, 100,250, 500, 1000, or 1500 nM UNG and 5OnM dU2AP-25-mer (Figure 1 6C). The average fluorescence at each UNG concentration was acquired and the data was fit to a non-linear regression curve. As shown in Figure 13, 2AP fluorescence increased as a linear function of UNG concentration until the binding/flipping reaction saturated between 500 and 1000 nM UNG. The net increase in 2AP fluorescence was obtained by subtracting the fluorescence of the enzyme acquired in the absence of dU'2AP-25-mer DNA from the total fluorescence intensity (Figure 16,legend).The saturation of 2AP fluorescence was not observed in binding reactions containing the R276 mutant enzymes. However, the 2AP fluorescence intensity observed at mutant enzyme concentrations from 0 to 500 nM did increase in a linear manner; therefore, the initial slope of the fluorescence binding curve as a measure of enzyme affinity for the diU- containing substrate. In the wild-type UNG reaction, the slope of the curve was determined from the 0, 100,250, and 500 nM UNG data points and found to be 22.7 (Figure 16C). In contrast, the defining slope of the 2AP fluorescence curve obtained for the R276E mutant was 0.134, 169-fold lower compared to that of UNG (Figure

1 6D).Using this approach, the relative binding affinities of the eighteen R276 mutant proteins for the dU-containing double-stranded DNA substrate were determined (Figure 16K). Mutations that resulted in the least 2AP fluorescence enhancement were R276D (0.00 1 %), R276P (0.08 %), and R276G (0.3 %), whereas the R276T (2.8 %), R276W (2.9 %), and R276S (3.9 %) mutations produced relatively more fluorescence enhancement. All R276X mutations negatively affected DNA binding! dijiU-flipping

(Figure16K). 123

Figure 13. Purity of the enzymes used in this study. A, UNG* (fraction VIII), UNG, and various Arg276 mutant proteins (fraction IV) were purified as described in Section 2.2.11. Samples (4 p.g) of UNG*, UNG, and each Arg276 mutant protein were subjected to 12.5% SDS polyacrylamide gel electrophoresis, and protein bands were visualized after staining with Coomassie Brilliant Blue G-250. The mobility of molecular weight standards (SDS-PAGE, low range standards, Bio-Rad) as well as the tracking dye (TD) are indicated by arrows from top to bottom, respectively (lane MWS). Lanes containing R276 mutant proteins are represented by conventional single letter amino acid abbreviations. B, UNG (fraction IV) was assayed for nuclease activity using 5'-end carboxyfluorescein-labeled single- and double-stranded oligonucleotide substrates, 5'-FAM-T-25-mer and 5'-FAM-TA-25-mer, respectively, as described in Section 2.2.12.3. Mock reaction mixtures contained 5'-FAM-T-25-mer (12.5 ng) or 5'-FAM-TA-25-mer (25 ng) in reaction buffer (lanes 1 and 8, respectively), control reactions contained 0.02, 0.2, or 2 units of E. coli exonuclease III and either 5'-PAM-T-25-mer or 5'-FAM-T'A-25-mer (lanes 2-4 and 9-11, respectively), and reactions containing UNG (8, 40, and 200 ng) and 5'-FAM-T-25- mer or 5'-FAM-T'A-25-mer (lanes 5-7, and 12-14, respectively) were incubated at 37 °C for 30 mm. The reaction products were resolved by 15 % polyacrylamide, 8.3 M urea, gel electrophoresis, and the gel was analyzed with a FMBioII fluorescence imaging system. Arrows indicate the locations of oligonucleotide substrate (3) and tracking dye (TD). 124

A. * Lane MWxiO3$.ACDEFGHILMNPQSTVWY 97.4L 66.2 - 45+ 31-. * ------. - * 21.5 14.4

TD

B. Lane 12345678 91011121314

S

iii]

Figure 13 125

Figure 14. Ability of UNG and Arg276 mutant proteins to bind Ugi. Reaction

mixtures (15j.tl)containing 40 pmol of E. coli Ung (lanes 2-3), UNG (lanes 4-5), or R276 mutant protein (lanes 6-4 1), with or without (+ or -) Ugi (100 pmol) were incubated as described in Section 2.2.12.4. A control reaction containing Ugi (100 pmol) alone was similarly processed (lane 1). Samples were analyzed at 4°C by non- denaturing 10% polyacrylamide gel electrophoresis, proteins were visualized with

Coomassie brilliant blue G-250stain,and the gel was imaged as described under Experimental Procedures. Arrows indicate the location of Ung, UNG, Ugi, UngUgi, UNGUgi, and the tracking dye front (TD). Arg276 mutant protein lane assignments are indicated by single letter amino acid abbreviations. 126

UngUNGA C D I Ugi:+ -+ -+ - +-+ -+ -+E -+F - G+ H L M Lane: 12 3 4 5 6 7 89 10111213141516171819 202122232425 Ung- - UNG UngUgi-.. - UNGUgi UgL. TD _$I

NP Q S I VW Y Ugi: - + + - +- +- +- +- + - + Lane:

- UNG * _ UNGUgi

I - -

Figure 14 127

Figure 15. Specific uracil-DNA glycosylase activity of R276X mutant proteins. Uracil-DNA glycosylase activity was measured under standard reaction conditions using 0.04-0.13 units of UNG*, UNG, or the indicated R276 mutant protein, as described inSection2.2.12.3. Protein concentrations were determined using the Bradford method (289) and the Protein Assay reagent (Bio-Rad). The specific activity (units/mg) of UNG* and the R276 mutant enzymes was normalized to that of UNG (3.76x io5units/mg), which was defined as 100 %. Arg276 mutant proteins are denoted by single-letter amino acid abbreviations. Error bars represent the standard deviation of four experimental determinations. 128

150 > >

C., < 100 C.,

C.) C) ci) 50 >C)

CC a,

UNG Mutant

Figure 15 129

Figure 16. Effect of Arg276 mutations on DNA-binding and base ifipping as measured by 2-aminopurine fluorescence. Reaction mixtures (250 p1) contained 50 nM double-stranded dNJU2AP-25-mer and various amounts (0-1500 nM) of UNG or Arg276 mutant proteins. Fluorescence intensity measurements were performed as

described inSection 2.2.17.A, The baseline fluorescent intensity of the 2- aminopurine-containing (dli,U.2AP-25-mer) DNA substrate was measured at 1 intervals for 1 mi UNG (500 nivIl) was then added and additional fluorescent intensity measurements were continued for 1 mm. An arrow (I) marks the time of UNG addition; zF1 is the average fluorescence intensity enhancement caused by TJNG addition. Lastly, Ugi (5 j.tM) was added to the reaction and the fluorescence intensity was monitored for another minute. The arrow II marks the time of Ugi addition and AF2 is the average 2AP fluorescent intensity quench that resulted from Ugi addition. The rectangular plot symbols represent the average intensity of 10 consecutive 1 s measurements. B, The fluorescent intensity of the dU2AP-25-mer was monitored for 1 mm as in (A) prior to the addition of Ugi (5 g.tM). The arrow III marks the time of Ugi addition; iF3 corresponds to the average increase in 2AP fluorescent intensity caused by Ugi addition. C, Reaction mixtures containing dU'2AP-25-mer (50 nM) and 0, 100, 250, 500, 1000, or 1500 tiM of UNG were prepared and the fluorescent intensity was determined at 1 s intervals for 1 mm. Open circles represent total observed fluorescence, open squares represent the fluorescence of the protein solution alone, and filled triangles represent the net fluorescence (open circles minus open squares). D, Samples were prepared as described in (C), except that the R276E mutant enzyme replaced UNG. The linear dependence of fluorescent intensity as a function of the enzyme concentration was calculated and termed the initial "slope" of the binding curve. Plot symbols are as described in (C). E, Sample preparation, fluorescent intensity measurements, and slope calculations were carried out as described in (D) for each of the Arg276 mutant enzyme preparations and compared to the initial slope of UNG, which was set to 100 %. Error bars indicating the standard deviation of the 60 130 is fluorescent intensity measurements in (C), (D), and (E) are obscured by the plot symbols.

22A. B.

18 T

14

F . 60 4-. 120 180 60 120 Time (sec)

r 25

15 a)

a) 5 C.) U) 0 500 1500 2500 UNG (nM) U-

5

3

I

500 1500 2500 R276E (nM)

100 -w> 90 .o 30 20 10 U- +c)AcoEFGH LMNPQSTVW UNG Mutant

Figure 16 131

3.1.5. Pliotochemical Crosslinking of UNG and Mutant Proteins to Single-stranded diy U-25-mer

To examine the effect of Arg276 mutations on UNG interaction with single- stranded DNA, UV crosslinking experiments with 5 '-end 32P-labeled single-stranded 25-mer oligonucleotide that contained a site-specific T, U, or dU deoxynucleotide were performed. The reaction mixtures were TJV-irradiated for 30 mm, subjected to non-denaturing polyacrylamide gel electrophoresis, and analyzed by Phosphorlmager as described in Section 2.2.13. As shown in Figure 17, two radioactive bands were identified: one band migrated rapidly and corresponded to free [32P]25-mer, while the other migrated more slowly and represented the UNGx[32P]25-mer protein-DNA crosslinked complex. The crosslinking efficiency of T.JNG to each oligonucleotides was: dU-25-mer (12 %)> T-25-mer (6 %) > U-25-mer (3%) (Figure 14A, lanes 6, 4, and 5, respectively). In order to determine the UV-dose dependence of UNGx[32P]25- mer crosslink formation, a time course of UV-irradiation was carried out using [32P]dijiU-25-mer DNA (Figure 17B). Inspection of the autoradiogram revealed the time-dependent emergence of the mobility retarded UNGx[32P]dNJU-25-mer crosslinked band (Figure 17B). Similar UV-irradiation time course experiments were conducted with [32P]T-25-mer and [32P]U-25-mer DNA. Quantification of the 32P radioactivity (Figure 17C) showed that, as before (Figure 17A), UV crosslinking was more efficient with [32P]dU-25-mer DNA than either [32P]T- or [32P]U-25-mer DNA. Moreover, the accumulation of UV-crosslinked UNGx[32P]25-mer complex increased in a linear maimer until approximately 20 miii, at which point no further increase in crosslinked product was observed for any of the DNA substrates (Figure 17C). Based on these results, the {32P]dU-25-mer DNA substrate and the 10 mm UV exposure time were selected for subsequent experiments. In order to ascertain the effect of the R276 mutations on DNA binding as reflected by UV crosslinking efficiency, each of the eighteen R276 mutant enzymes were combined with [32P]dU-25-mer DNA, irradiated for 10mm,and the crosslinking efficiency relative to UNG was determined 132

as described above (Figure 14D). Mutations that resulted in the lowest crosslinking efficiency were R276E (16 %), and R276G (13 %), whereas the R276C (96 %) and R276L (94%) were essentially same as UNG. Mutation of Arg276 to an aromatic amino acid resulted in somewhat reduced crosslinking efficiencies, as R276F, R276H, R276W and R276Y exhibited 62, 71.5, 62, and 65 %, respectively, of the UNG crosslinking efficiency. Mutation of Arg276 to amino acids with aliphatic side chains (R276A, R2761, and R276V) resulted in 58, 54, and 44 % of UNG crosslinking efficiency, respectively; however, as stated above, R276L exhibited 94 % efficiency, whereas R276G exhibited 13 %. Mutation of arginine 276 to amino acids with sulfur or hydroxyl-containing side chains (R276M, R276S, and R276T) resulted in 62, 39, and 43% of UNG crosslinking efficiency; the exception, as noted above, was R276C (96 %). Finally, acidic amino acid substitutions (R276D, R276N, R276Q, and R276E) and substitution with proline (R276P) resulted in 69, 57, 42, 16, and 48 % of UNG crosslinking efficiency, respectively (Figure 1 7D). In conclusion, all R276 mutations, except R276C and R276L, showed a decrease in UV crosslinking efficiency.

3.1.6. Processivity of UNG on Concatemeric Uracil-containing [32PJDNA

Proteins that recognize a particular site or sequence in DNA can locate their targets by random three-dimensional diffusion or by facilitated diffusion (72), processes often referred to, respectively, as distributive or processive search mechanisms. To determine the search mechanism used by UNG to locate sequential uracil residues on the same DNA strand, a defined concatemeric DNA substrate was constructed from repeating 30 nucleotide monomer units (Figure 18A), as described in Section 2.2.14. Each monomer unit contained a UA base pair in a defined sequence context; the uracil-containing strand was 5'-end 32P-labeled. Thus, a [32P]29-mer bearing a 3'-phosphate ([32P]29p-mer) would be released if adjacent uracil residues were successively excised (processive excision) and the apyrimidinic sites cleaved by alkali treatment. A [32P]29p-mer would be released in the event of successive uracil 133

Figure 17. Abffity of UNG and Arg276 mutant proteins to form UV-catalyzed

cross-links to [32P]25-mer DNA. A, Samples (12.tl)containing 20 pmol of 5'-end 32P-Iabeled oligonucleotide T-25-mer, U-25-mer,or dNJU-25-mer without UNG (lanes 1-3, respectively) or with 40 pmol of UNG (lanes 4-6, respectively) were UV- irradiated for 30 mm as described in Section 2.2.13. Following irradiation, samples were subjected to non-denaturing polyacrylamide gel electrophoresis; the gels were dried and analyzed using a Phosphorimager. The positions of the UNG [32P]DNA-25- mer cross-linked complex bands and free [32P]DNA-25-mer) bands are indicated by arrows. B, Reaction mixtures (12 j.d) were prepared in duplicate that contained 20 pmol of {32P]dU-25-mer and 40 pmol of UNG. Following UVirradiation for 0, 5, 10, 20, 30, and 45 mm (lanes 3-14, respectively), reactions were analyzed in (A). Control reactions (lanes 1, 2) containing 20 pmol of [32P]dU-25-mer, were not irradiated. C, Reaction mixtures (12 j.il) were prepared in duplicate that contained either 20 pmol of [32P}diijU-25-mer (closed circles), [32P]T-25-mer (closed squares), or [32P}U-25-mer (closed triangles), and 40 pmol of UNG. Following UV-irradiation for 0, 5, 10, 20, 30, and 45 miii, reactions were analyzed as in (A), and the Phosphorlmage data were quantified using the ImageQuant program. The cross-linking efficiency (%) was calculated by dividing the intensity of the IJNGx[32P]25-mer band by the sum of the [32P]25-mer and UNGx[32P]25-mer bands and multiplying by 100. B, Reaction mixtures (12 tl)were prepared that contained 20 pmol of [32P]d'qiU-25-mer and 40 pmol of each Arg276 mutant enzyme. The reactions were UV-irradiated for 10 miii and analyzed as described in (C). The cross-linking efficiency of each mutant preparation, indicated by the corresponding single letter amino acid abbreviation, is compared to that of UNG. Error bars represent the standard deviation of three experiments. 134

A. B. 123456 1234567891011121314

UNGx25.m x iU-25mer

25arj I ,,,Øs*.øsø* iiU-25-mer

20 c. .9 C., 15 uJ

10

10 20 30 40 50 Time (mm)

C) C 140

oQ100 '-C 0.9 w 60

20

UNG Mutant

Figure 17 135

cleavage rather than a [32P]30-mer, since hot alkali catalyzes 3-elimination of the 3'- phosphodiester bond of the AP-site followed largely by 3-e1imination of the 5'- phosphodiester bond of the f3-elimination product, resulting in the loss of the baseless sugar. However, if uracil residues were excised at random, little [32P]29p-mer would be produced. By comparing the amount of [32P]29p-mer released relative to the total amount of uracil excised during the early stages of the reaction, the degree of processive excision (processivity) of each enzyme could be evaluated. This approach was successfully used previously to determine the impact of ionic strength on the processive search mechanism of E. coli uracil-DNA glycosylase and rat mitochondrial uracil-DNA glycosylase (62). In the present report, the 5'- and 3 '-ends of the [32P]U'A- 30-mer monomer unit contained an eight-nucleotide complementary overlapping sequence that formed an HpaII restriction endonuclease recognition site upon hybridization of two or more monomer units (Figure 1 8A). Each [32P]U.A-30-mer monomer also contained an internal EcoRJ cleavage site. Concatemeric [32P]DNA substrates with uracil residues at regular intervals (30 nucleotides) along one strand of the duplex polymer were produced by DNA ligation. Analysis of the ligation products by non-denaturing polyacrylamide gel electrophoresis revealed that greater than 90 % of the polymeric [32P]U"A DNA was -P300 bp. In order to isolate high molecular weight [32P]U.A concatemeric DNA, the polymeric [32P1U.A ligation products were subjected to gel filtration chromatography, and fractions that contained [32P]U.A concatemeric DNA greater than 600 bp as judged by non-denaturing polyacrylamide gel electrophoresis were pooled and used as substrate for processivity reactions The average length of the concatenated [32P]UA DNA substrate was determined by three additional methods, all of which involved enzymatic "collapse" of the [32P]U.A concatemeric substrate to its monomer components (Figure 1 8A). First, the concatemeric substrate was subjected to excess TJNGand alkali treatment, and the percentage of [32P] 1 4p-mer determined following denaturing polyacrylamide gel electrophoresis (Figure14B, lane2). The ratio of [32P] 1 4p-mer to the sum of [32P]29p-mer plus [32P] 1 4p-merwas dictated by the length 136

of the concatemeric substrate, since digestion of every uracil residue released 32P- containing 29p-mers except at the 5'-and 3'-ends. The 5'-end digestion fragment was a [32P] 1 4p-mer; however, the 1 5-mer 3'-end fragment did not contain [32P] and therefore was undetectable. UNG digestion and subsequent alkaline treatment was found to produce 5.6%[32PJ14p-mer. In the second and third methods, the length of the average [32P]concatemer was determined by restriction endonuclease digestion (Figure 1 8A). Digestion of the [32P]U.A concatemer with Hpa II produced [32P]30-mers except at the 5'- and 3'- ends, which released [32P]28-mer and [32P]32-mer, respectively (Figure 18B, lane 3). The amount of [32P]28-mer was 9.2 %, whereas the amount of [32P]32-mer was 9.3 %. Digestion of the [32PU'A concatemer with EcoPJ yielded [32P]30-mer except for the 5 '-end, which released [32P]20-mer; the percentage of [32PJ20-mer was 6.2 % (Figure 1 8B, lane 4). While the amounts of 5'-end [32P]fragments generated by the UNG and EcoPJ digestionswere in good agreement (5.6 and 6.2 %, respectively), the amount of the 5'-end [32P]28-mer generated by HpaII digestion of the concatemeric [32P}DNA substrate was somewhat higher (9.2 %). Therefore, the data from the UNG and EcoRJ digestions was used to estimate the length of the average concatemeric [32P]DNA substrate, which was found to be approximately 510 bp. As shown in Figure 18B (lanes 2, 3 and 4, at the S arrow) it appeared that a minor percentage of the concatemeric DNA substrate was intractable to UNG (<7.4 %), Hpa II (<0.8 %) or EcoR I (<0.2 %) cleavage. A minor band at the 60 nt position that was refractory to cleavage was also observed. However, analysis of subsequent processivity experiments was not materially affected by the minor refractory DNA substrates. The estimate of the average size of the [32Pjconcatemeric DNA substrate as determined by non-denaturing polyacrylamide gel electrophoresis was higher than the estimates provided by the method of enzymatic collapse, since annealed but unhigated junctions between monomer units would not be detected during non-denaturing polyacrylamide gel electrophoresis. In order to measure the processivity of UNG, it was critical to relate the amount of processive excision product ({32P]29p-mer) detected to the total amount of 137

uracil excised. The [32P]concatemeric DNA substrate was designed such that following uracil excision by UNG, exhaustive digestion withHpaIIor EcoRJ, and hot alkali treatment, would produce [32PJ30-mer if the uracil residue within the monomer unit wasnotexcised, and {32P] 16p-mer and [32P]24p-mer, respectively, if the uracil

residue was excised (Figure18B, lanes 5and 6). Thus, the extent of UNG digestion (the overall amount of uracil excised during the reaction) could be determined, using EcoRI for example, as the ratio of [32P]24p-mer detected to the sum of [32P]24p-mer + [32PJ30-mer detected (Figure 1 8A, Reaction 5). Processivity could then be assessed by plotting the amount of processive product produced ([32P]29p-mer) as a function of the extent of digestion. Since processive excision generates more [32P]29p-mer, the ratio of [32P]24p-mer to [32P]24p-mer+[32Pj30-mer would be greater than that generated by distributive excision, which would produce little [32P]29p-mer, especially during the early stages of the reaction. In the case of distributive excision, the accumulation of [32P]29p-mer products in large amounts would appear only late in the reaction time course when a significant percentage of the uracil residues on a particular [32P]concatemer had already been excised, raising the likelihood thata single distributive excision event would release a [32P]29p-mer. To assess UNG processivity, a high molar ratio of DNA substrate to enzyme was used to ensure that, on average, no more than one UNG molecule was bound initially to a particular concatemeric-DNA molecule. Standard processivity assays were performed in reactions containing 10 pmol of concatemeric [32P]U.A DNA substrate and 0.06 units of UNG. A time course reaction of uracil excision was conducted, and the products of each reaction time point were divided into two aliquots. One aliquot was subjected to AP-site hydrolysis and used to determine the percentage of total [32P]DNA substrate that appeared as {32P]29p-mer units that were indicative of a processive excision search mechanism. The other aliquot was used to determine the extent of digestion, i.e., the amount of uracil released from polymeric [32P]DNA substrate regardless of position. The extent of digestionwas measured by treating the UNG reaction products with excessEcoR Irestriction endonuclease, as 138

described above. Both aliquots were then analyzed together by 12 % denaturing polyacrylamide gel electrophoresis (Figure 18C). As the reaction progressed, only the [32P}29p-mer productswere visible and partially digested {32P]oligonucleotides

appeared at the relatively low level in the later time points (Figure 1 8C,lanes1-7). The extent of UNG digestion at each point in the reaction time course was evaluated by

EcoPJ digestion Figure 1 8C,lanes8-14). Quantification of the gel results revealed that the emergence of the [32P]24p-mer bands appeared commensurate with the production of the [32P]29p-mer bands in the UNG reaction product, as shown in Figure 16C. Similar results were found in reactions containing either 0.04 units of Ung or 0.06 units of UNG* and the same 32P-labeled concatemeric DNA substrate (Fig. 16A and B, respectively). The molar quantities of [32P]29p-mer excision products generated by Ung and UNG* on the concatenated uracilcontaining [32P]DNA were dominant throughout the reaction time course. Analysis of the Ung and TJNG* gels also showed that the emergence of the [32P}24p-mer band was proportional to the production of the [32P]29p-mer band in both Ung and UNG* reactions (Figure 1 9C). To arrive ata more quantitative assessment of processivity, the amount of processive excision product ([32P]29p-mer, Y axis)was divided by the extent of digestion (X axis) at each early time point to determine the initial slope of the "processivity" plot. The slope of the E. coli Ung processivity plot was determined to be 0.51, which was in good agreement with an earlier measurement (0.56) of Ung processivity reported by Bennettet al.(62). The slope of the UNG* processivity plot was determined to be 0.44, consistent with an earlier processivity measurement (0.45) reported for rat liver mitochondrial uracil- DNA glycosylase (62), and the slope of the UNG processivity plot was 0.39. These results suggested that the core catalytic domain of human uracil-DNA glycosylase (l.JNG*), like E. coli Ung and rat liver mitochondrial uracil-DNA glycosylase, located uracil residues using a processive search mechanism. In addition, the results indicated that the N terminal His6-tag of UNG did not significantly affect the processive character of the enzyme, as the processivity analysis of the UNG* and UNG reaction products were very similar (Figure 19(T). 139

3.1.7. Processivity of R2 76 Mutants on Concatemeric Uracil- containing [32PJDNA

To assess the effect of Arg276 mutations on the processive search mechanism of UNG, the following four mutants were selected for analysis: i) R276L, chosen since the efficiency of UV cross-linking to [32P]IJ-25-mer was unaffected; ii) R276W, since it had highest activity among aromatic side chain mutants; iii) R276D and iv) R276T were chosen because they exhibited similar moderate defects in uracil-DNA glycosylase activity and UV crosslinking efficiency. The standard processivity assay was conducted in reactions containing 10 pmol of concatemeric [32P]U'A DNA substrate and various amounts of R276D, R276L, R276T, and R276D as described in the legend to Figure 19. The reaction products of each time point were analyzed for processive [32PJ29p-mer product release and the extent of digestion as described above. Next, the processivity of the mutant proteins was assessed by graphing the percentage of processive [32PJ29p-mer product released as a function of the extent of digestion, and analysis of the results showed that the release of [32P]29p-mer was proportional to the overall amount of uracil excised. (Figure 1 9D). From the processivity plots the initial slopes were determined for R276D, R276L, R276T, and R276W, and found to be 0.39, 0.41, 0.55, and 0.54, respectively. These results were in general agreement with those obtained for UNG, UNG*, and Ung (Figure 1 9C). Moreover, similar results were observed for processivity reactions containing R276H, R276C, and R276S. Taken together, these experiments indicated that mutations at Arg276 did not significantly affect the processive search mechanism of UNG. 140

Figure 18. Analysis of reaction products generated by UNG from a concatenated uracil-containing [32P]DNA substrate. A, Concatenated [32P]DNA substrates were prepared by ligation of double-stranded 30-mer units as described in Section 2.1.14. Each 30-mer unit contained a uracil residue at position 15 in the upper strand opposite adenine in the lower, complementary, DNA strand; only the uracil-containing strand was 5'-end 32P -labeled (*). An EcoRI restriction endonuclease recognition site was situated 3' to nucleotide 20 of the uracil-containing DNA strand. The 30-mer unit was designed with self-complimentary overhanging 5'- ends of eight nucleotides that formed a HpaII endonuclease recognition site upon ligation. Ligation of the 30-mer units created a concatenated uracil-containing [32PJDNA substrate in which each uracil residue as well as the EcoRI and HpaII recognition sites were separated by 30 nucleotides. Reaction 1: Sequential uracil excision by UNG followed by alkaline hydrolysis of the resultant AP-site produces 29-mers containing an internal 32P-label and a 3'-phosphate (29p-mer). Excision of the 5'-terminal uracil produces a 5'-end 32P- labeled 14-mer with a 3'-phosphate (14p-mer). Reaction 2: HpaII digestion of the concatenated [32P]DNA substrate and subsequent alkaline hydrolysis generates 30- mers with an internal 32P-label. Cleavage by HpaII at the 5'- and 3'- ends yields a 28- mer containing a 5'-end 32P-label and a 32-mer containing an internal 32P-label, respectively. Reaction 3: EcoRI digestion produces 30-mers except at the 5'-terminus where a [32P]DNA-20-mer is generated. Reaction 4: Concatenated [32P]DNA partially digested with UNG, then exhaustively digested with EcoRI followed by alkaline hydrolysis, yields 24-mers containing a 3'-phosphate (24p-mer) where the uracil has been excised, but [32P30-mers from uracil-containing 30-mer units, as shown in Reaction 3. Only 32P-labeled reaction products are shown. B, Characterization of the concatenated uracil-containing [32P]DNA substrate. Reaction mixtures (10 tl) containing 10 pmol of concatenated {32P}DNA and either no addition (lane 1), 70 units of UNG (lane 2), 10 units of EcoRI (lane 3), 10 units of HpaII (lane 4), 70 units of 141

UNG and 10 units of HpaII (lane 5) or 70 units of UNG and 10 units of EcoRI (lane 6), were incubated at 37 °C, subjected to alkaline hydrolysis, and analyzed byl2 % denaturing polyacrylamide gel electrophoresis as described in Section 2.2.3.3. The mobility of the unreacted concatenated [32P]DNA substrate (S) and various oligonucleotides reaction products (32-, 30-, 29p-, 28-, 24p-, 20-, l6p- and 14p-mer) is indicated by arrows. C, Time course of UNG digestion. A reaction mixture (100 p.1) containing 0.06 units of UNG and 80 pmol of concatenated uracil-containing {32P]DNA was incubated at 37 °C. Aliquots (10 p.!)were removed at 0, 5, 10, 20, 40, and 80mmand terminated by the addition of Ugi as described in Section 2.1.16. A control reaction contained 10 pmol of the [32P]DNA substrate only. Following reaction termination, each time point was divided into two aliquots. One, designated the UNG reaction, was subjected to alkaline hydrolysis. The other, designated the EcoRI reaction, underwent EcoRlI digestion and subsequent alkaline hydrolysis. Reactions were analyzed by denaturing 12 % polyacrylamide, 8.3 M urea gel electrophoresis as described under Experimental Procedures. UNG reactions corresponding to the control reaction, and 0, 5, 10, 20,40, and 80mmtime points were loaded in lanes 1-7, respectively, whereas the corresponding EcoRI reactions were loaded in lanes 8-14. Arrows indicate the positions of the various oligonucleotide reaction products. The extent of uracil excision after each incubation time was determined by dividing the amount of [32P]24p-mer oligonucleotide released by EcoRJ digestion by the sum of [32P]24p-mer+[32P]30-mer, and multiplying the quotient by 100; this percentage was called the Extent of Digestion. 142

A. TJNG EcoRI HpaII

GTGACTATTcACTTAAGCTGGCCGCACTCG DNA Ligase

U++*Ut+*H Ut+* U*** Ut**H Ut 3' 5' I Reaction 1. UNG ner [29mer']x

u * U 2. HpaII *U 1* U * 28-mer 1.30-mer3 32-mer 32-mer 3. EcoRl 20-mer [O-m'r'x

4. UNG + EcoRI * [ * U 24-mer 30-mer3x B. C. Lane: 1 234 51 2 3 4 5 6 7 8 9 10 1112 13 14 s-.i-' I_'

nt I- 120 I90 -

0*4 4 . ' ! 24p

Figure 18 143

Figure 19. Analysis of reaction products generated by Ung, UNG*, UNG or its Arg276 mutant proteins from concatenated uracil-containing [32P]DNA. Two

processivity reaction mixtures (100j.tl)containing either0.04units of E. coli Ung (A) or 0.06 units of human UNG* (B), and 80 pmol of concatenated uracil-containing

[32P]DNA substratewere prepared as described in Section 2.2.14. Samples (10j.tl)

were withdrawn after 0, 5, 10, 20,40,and 80 miii of incubation at 37 °C, and divided into two aliquots as described in the legend to Figure 7. From the reactions subjected to alkaline hydrolysis only, the molar quantities of the four detectable [32P]oligonucleotide UNG digestion productswere determined and are presented graphically as: 29p-mer (black bar), 59p-mer (cross-hatched bar), 89p-mer (striped bar), and 11 9p-mer (wavy bar). C, The amount of [32P]29p-mer product generated at each incubation time (A and B, black bars) was determined relative to the total [32P]DNA detected, and the extent of digestionat various incubation times was determined after EcoRI treatment. A time course reaction identical to that described for Ung and T.JNG* but containing 0.06 unit of UNG was also conducted and analyzed as described above. The processivity of Ung (0.04 unit, filled circles), UNG* (0.06 unit, filled squares), or UNG (0.06 unit, filled triangles) was analyzed by graphing the amount (%) of processive [32P]29p-mer product detected as a function of the extent of digestion (%). D, Processivity analysis of uracil-DNA glycosylase reactions containing R276D (0.3 unit, open diamonds), R276L (0.3 unit, open circles), R276T (1.0 unit, open triangles), and R276W (1.0 unit, open squares) is presented and compared to UNG (0.06 unit, filled circles). 144

0

0

2 a- N -a

Time(mm)

80 C.

60

40

20 0 I- 0 80 D.

C

20 40 60 80 100 Extent of Digestion (%)

Figure 19 145

3.2. Discussion

The negative effect of R276 mutations on UNG catalytic activity may have resulted from perturbation of the enzyme-DNA interaction during the formation of the enzyme-DNA complex prior to glycosylic bond cleavage. Wong and co-workers (121) used pre-steady state kinetic analysis of a single catalytic turnover by E. coli Ung to predict three sequential steps in the enzyme mechanism that occur prior to cleavage of the glycosylic bond: 1) rapid equilibrium binding of Ung to DNA (non-specific), 2) specific uracil binding and flipping, and 3) protein conformational change. The possibility that mutation of Arg276 affects the initial binding step of UNG to double- stranded DNA will be discussed in light of three experimental observations. First, structural data from the double L272R1D145N UNG mutant in complex with (cleaved) UG DNA indicated that the g-NH2 of the Arg276 guanidinium side chain formed a hydrogen bond with the oxygen residue of the third phosphate residue 3' to the site of the cleaved uracil residue (115). Also, in the co-crystal structure of wild-type UNG in complex with either cleaved UG or cleaved U'A DNA, the Arg276 side chain ii- nitrogen was reported to form a water-bridged hydrogen bond with theN3of an adenine residue 3' to the uracil residue, aninteraction described as "reading" the DNA minor groove (116). Thus, mutations at Arg276 may disrupt UNG-DNA interactions, effectively lowering the affinity of the enzyme for DNA. However, it is not at all clear that disruption of one positively charged amino acid side chain out of thirteen would strongly affect DNA binding, since it appears that electrostatic interactions, not direct or water mediated DNA phosphate interactions, orient the enzyme on DNA for initial damage detection (116). It should also be noted that the

Arg276-DNA phosphate interaction observed by Slupphauget al.(115) was not reported by Parikhetal.(116). To determine the mechanism of catalysis by UNG, Dinner and co-workers (128) used a hybrid quantum-mechanicallmolecular- mechanical approach based on the UNG-DNA co-crystal structure solved by Parikhet al.(117). They calculated that the primary contribution to lowering the activation 146

energy of uracil excision came not from the enzyme but rather from the burial and positioning of four 5'-phosphate groups in the substrate (T4, dU5, dA6, and dT7) that were thought to stabilize the cationic sugar in the transition state. According to this model, an interaction between Arg276 guanidinium side chain and the 5 '-phosphate of dC8or lack thereof would not significantly affect the activation energy of uracil excision. Second, the fluorescence equilibrium binding experiments reported in this manuscript demonstrated that binding of R276 mutant proteins to a dU2AP- containing duplex oligonucleotide was not saturable. This result suggested that the Li of the R276 mutant proteins for duplex DNA was significantly elevated relative to wild-type UNG. However, in these experiments it was not possible to distinguish between binding to DNA and binding the flipped-out uracil. The fluorescence signal in these experiments came from DNA containing 2AP opposite wU; flipping of the di1iU residue led to increased 2AP fluorescence. Therefore, it may be argued that the R276 mutations affected the efficiency of dU-flipping rather than DNA binding. Parikhet al. (116) observed that mutation of Leu272 to Ala did not affect the Tate of binding

(kass) to 4'-tbio-UG DNA relative to wild-type IJNG, but the rate of dissociation(kdi5s) was approximately three times slower for the wild-type enzyme, presumably because insertion/retraction of the Leu272 side chain in the DNA minor groove slowed dissociation from the DNA phosphate backbone. In Wong's model (121), insertion of the Leu272 loop acts as a "door stop" to prevent facile return of the flipped out uracil to the DNA base stack. Structural data from Parikhetal. (116) indicated that the cN of the Arg276 guanidinium side chain also participated in a water-mediated hydrogen bond with the carbonyl oxygen of Leu272. Therefore, ablation of the Arg276 side chain by mutation could destabilize the Leu272 intercalating ioop and reduce its ability to prevent the uracil residue from flipping back into the DNA base stack. In addition, mutation of Arg276 may interfere with the ability of Tyr275 to widen the DNA minor grove and facilitate uracil flipping (116). 147

Third, the effect of mutations at Arg276 on UV-catalyzed crosslinking to single-stranded thjiU-containing oligonucleotide was much less severe than the effects

observed for experiments involvingdouble-strandedDNA. In fact, the UV crosslinking efficiency of two mutant proteins, R276C and R276L, was comparable (95 %) to that of wild-type. This result indicated that while Arg276 was required for efficient recognition and base flipping of uracil in double-stranded DNA, its role in binding the more flexible single-stranded 25-mer DNA was much less important. The efficiency of UV crosslinking UNG to dNJU-25-mer was approximately four-fold greater than to U-25-mer and two-fold greater than to T-25-mer. This result is

consistent with the report by Shoyeret al.(124), who observed that UV crosslinking of E. coil Ung to single-stranded DNA 20-mer containing one uracil residue amid 19 dTMP residues was less efficient (-' 0.5) than to dT2o. It is likely that UV crosslinking of UNG to dijiU-25-mer was more efficient due to the non-cleavable nature of the Cl- Cl' glycosylic bond linking the pseudouracil base to deoxyribose. In this respect, the residence time of flipped dNJU in the active site may be longer, and the association of UNG active site residues with dU and adjacent nucleotides prolonged. If the duration of the protein-DNA contact were longer, then the efficiency of UV crosslinking would be expected to increase, as was observed. While every amino acid side chain has the potential to form UV-catalyzed crosslinks with DNA, the aromatic side chains of Phe, Trp, Tyr, and His were reported to form DNA crosslinks more efficiently (293). However, substitution of these aromatic amino acids for Arg276 did not lead to greater IJV crosslinking efficiency; instead, substitution with Cys or Leu resulted in near wild-type crosslink formation. While a straightforward explanation for these observations is not obvious, it is possible that the differential TJV crosslinking efficiencies of the R276 mutant proteins may have to do with the relative DNA binding affinity for the single-stranded DNA. Using a concatemeric [32P]U.A polynucleotide substrate with uracil residues and anEcoRlendonuclease recognition site at non-overlapping 30 nucleotide intervals, the catalytic domain of human uracil-DNA glycosylase was found to utilize 148

a processive mechanism to excise successive uracil residues. This result was consistent with previously reported processivity measurements of E. coli Ung and rat liver mitochondrial uracil-DNA glycosylase (62).The interpretation of a processivity plot slope measurement of 0.5 means that, on average, two uracil residues are excised per DNA encounter, where, for a slope p, the expected number of excisions is 11(1 -p) (R. T. Smythe, Oregon State University, personal communication). Thus, uracil-DNA glycosylase from different sources exhibit modest processivity, consistent with short sliding or hopping facilitated diffusion (294). Interestingly, the processivity of the R276X mutant proteins studied did not differ significantly from that of UNG, although the Arg276 mutations resulted in uracil-DNA glycosylase enzymes with notably decreased catalytic activity. Thus, aggregate electrostatic interactions between the DNA binding groove of the enzyme and non-specific DNA may be the primary determinant of the search mechanism. 149

4. Mutational Analysis of Arginine 276 in the Leucine-loop of Human Uracil-DNA Glycosylase (II)

As discussed in the previous chapter, the structural data of UNG* in complex with DNA (Figure 12) suggested that the role(s) of Arg276 might involve 1) enzyme- DNA-phosphate interactions, and 2) stabilization of the Leu272 side chain, either before or after it was inserted into the DNA minor groove. The first issue was investigated in Chapter 2 of this dissertation. In Chapter 4 studies are presented that address the role of Arg276 in facilitating leucine-loop insertion, stable uracil binding, enzyme conformational changes, and catalytic activity. The transient kinetics were employed to study the effect of six different amino acid substitutions at Arg276 (R276C, R276E, R276H, R276L, R276W, and R276Y) on the enzyme conformational change associated with binding uracil in double- and single-stranded DNA. The rationale for choosing these particular Arg276 mutant proteins has been described in Section 3.1.7. The catalytic activity of these mutant enzymes was determined using single-and double-stranded uracil-containing oligonucleotide substrates in 10 mm reactions. The results presented here demonstrate that Arg276 was required for efficient processing of uracil located in double-stranded DNA, but dispensable for excision of uracil in single-stranded DNA.

4.1. Results

4.1.1. Fluorescence Properties of dtiU2AP-25-mer

2'-deoxypseudouridine (diiU) is similar in structure to 2'-deoxyuridine and forms a Watson Crick base pair with deoxyadenosine in duplex DNA (295). However, dU is different from deoxyuridine in that the pseudouracil base is linked to deoxyribose by a C-glycosylic (Cl-Cl') bond rather than by the N-glycosylic (Nl-Cl') bond found in deoxyuridine. The Cl-Cl' bond can not be cleaved by the core catalytic 150

domain of human uracil-DNA glycosylase (UNG), and provides an uncleaved transition-state-like UNG-DNA complex (117). In order to investigate the DNA binding affmity of each Arg276 mutant enzyme, an oligonucleotide containing a fluorescent analogue of adenine, 2-aminopurine (2AP), was used. The fluorescence properties of 2AP have been used to probe the variations in DNA structure, DNA glycosylase-mediated base flipping, and protein-DNA interactions (12 1,296-300). The intrinsic fluorescence of 2AP is quenched when the base analog is incorporated into duplex DNA (118,298). The reduction in 2AP fluorescence in duplex DNA is attributed to stacking interactions with neighboring nucleobases (298). In order to study the effect of R276 mutations on UNG-DNA interactions, the properties of dU and 2AP were utilized and a duplex dNiU2AP-25-mer was constructed, in which a site-specific dU residue was positioned opposite 2AP. In order to verify the usefulness of the duplex dNJU2AP-25-mer, the fluorescence properties of dNJU.2AP-25-mer were examined. The observed emission spectrum (Xex = 310 mn) of 400 nM solution of 2AP-25-mer showed incremental fluorescence quenching upon addition of increasing concentrations of the complementary, dU- containing oligonucleotide (dU-25-mer) up to an equimolar concentration of 400 nM (Figure 20B, a-f, respectively). Addition of excess dU-25-mer did not result in further fluorescence quenching (Figure 20B, g-h). Fully duplex dU2AP-25-mer exhibited characteristic 2AP excitation and emission maxima of 310 and 370 nm, respectively (Figure 20A,arrows).The 2AP fluorescence intensity of dU2AP-25-

mer at 370 nm (Xe,,= 310 nm) remained unchanged during a 10 mm time trace indicating that 2AP fluorescence was stable. Thus, the correlation of 2AP fluorescence quenching with dU2AP-25-mer duplex DNA formation and the stability of 2AP fluorescence were consistent with previous reports (121,301). 151

4.1.2. Effect ofArg2 76 Mutations on UNG-DNA Interactions

In Chapter Three, mutations at Arg276 were found to attenuate 2AP fluorescence intensity when the mutant enzymes bound diU.2AP-25-mer (Figure 16). In order to quantitatively determine the dissociation constant (Kd) of UNG and the R276 mutant proteins for binding dijU'2AP-25-mer, six R276 mutant enzymes, together with wild-type UNG, were selected for further analysis: R276C, R276L, R276H, R276E, R276W, and R276Y. These six mutant proteins were chosen based on findings described in detail in Chapter Three: i) the efficiency of R276C and R276L T.JV-crosslinking to [32P]dU-25-mer was comparable to wild-type UNG; ii) the activity of R276H and R276E was highest and lowest, respectively, of the eighteen mutant enzymes; and iii) defects in the catalytic activity of the aromatic R276W and R276Y mutant proteins were moderate, and the fluorescent properties of the R276W side chain might serve to report on changes in the Arg276 environment. Equilibrium binding measurements were conducted using dNJU'2AP-25-mer (100 nM) and increasing amounts (0-2 tM) of UNG. The average fluorescence at each UNG concentration was acquired and the data was fit to a non-linear regression curve (see below). The net increase in 2AP fluorescence was then obtained by subtracting the fluorescence of the enzyme acquired in the absence of dU2AP-25-mer DNA from thetotalfluorescence intensity. As shown in Figure 21A, 2AP fluorescence increased as a linear function of UNG concentration and saturated at approximately 500 nM UNG. This 2AP fluorescence enhancement was dependent on UNG DNA- binding activity, since pre-treatment of UNG with the uracil-DNA glycosylase inhibitor protein, Ugi, totally abolished the 2AP fluorescence enhancement, as demonstrated in Figure 16. The net fluorescence enhancement (filled circles) at each UNG concentration (Figure 20A,fihled circles) was then fitted to Equation (1), and the Kd value for UNG binding to dU2AP-25-mer was determined to be 185 ±7 nM. Similarly, when dU2AP-25-mer (100 nM) was titrated with increasing amounts (0-2 j.tM) of R276C, a concentration-dependent increase of 2AP fluorescence was also 152

observed (Figure 21B). However, the 2AP fluorescence curve saturated at -4500 nM R276C, and the magnitude of 2AP fluorescence enhancement at saturation was significantly reduced as compared to that of UNG (Figure 21B). The R276C net fluorescence enhancement data were fitted to Equation (1) and the K value for R276C binding to dijU2AP-25-mer was determined to be 490±49 nM. Therefore, the reduction in 2AP fluorescence intensity by R276C correlated with an increased K1. Identical dU bindingl2AP fluorescence enhancement experiments were conducted for the R276E, R276H, R276L, R276W, and R276Y mutant proteins, and theKD values determined were 1428, 686, 945, 824, and 799 nM, respectively. As shown in Figure 21 C, the R276 mutant proteins displayed a 2.6- to 7.7-fold reduction in DNA binding affinity for dU2AP-25-mer compared to UNG; R276E exhibited the largest reduction in binding affinity.

4.1.3. Pre-steady State Kinetic Analysisofthe UNG Fluorescence Change Associated with Binding Duplex Uracil-DNA

Crystal structures of UNG alone and in complex with the non-hydrolyzable thU-containing substrate indicated that UNG underwent a global conformational change upon binding uracil in DNA (117). The change in conformation was determined by comparing the structure of the free enzyme to the structure of the enzyme in complex with dU-DNA, and the transition was described as that from an unbound, open conformation to a uracil-DNA-bound, closed conformation (Figure 22) (115,116). Binding involved compression by DNA binding loops and the closure of a central gap between 131 and 133 that extended hydrogen bonding interactions by two residues, allowing inter-j3-strand H-bonds, and was termed a 13-zipper (113,114). These movements effectively clamped the enzyme around the flipped-out uracil nucleotide, creating the enzyme active site by bringing key residues into functional position, and inserted the leucine-intercalation loop into the DNA minor groove to hinder return of the uracil nucleotide to the base stack (117). In pre-steady studies of 153

DNA damage recognition and uracil flipping by E. coli uracil-DNA glycosylase (Ung), Stivers et al. (118) observed transient quenching of Ung intrinsic tryptophan fluorescence associated with uracil flipping. The authors proposed that the fluorescence quenching reflected an enzyme conformational change associated with the leucine-loop insertion step (118). To determine whether a change in UNG conformation could be detected in real time, UNG was mixed with excess dsU.A-25-mer DNA and the intrinsic protein fluorescence was monitored. The kinetic trace of protein fluorescence showed a rapid fluorescence quenching in the first 15 milliseconds followed by a slower fluorescence recovery over 2 seconds (Fig. 23B). This kinetic pattern was uracil-DNA-specific, since no fluorescence change was observed when UNG was reacted with dsTA-25- mer DNA (Fig. 23A). Hence, the change in UNG intrinsic fluorescence upon uracil- DNA binding most likely reflected the structural transition of UNG from an open conformation to a closed, uracil-DNA-bound conformation. Since UNG, like Ung, contains several (7) tryptophan residues, the change in enzyme conformation decreased the fluorescence of one or more of these residues (Figure 24A). Although the identity of the Trp residue(s) involved in fluorescence quenching has not been determined, it is possible that Trp222 and (or) Trp245 may be responsible, since

Trp222 is close to11 and Trp245 is located at the active site-end of 13 (Figure 24B). Stabile uracil binding is accompanied by closing of a central gap between j31 and 3 (the 3-zipper), which extends 3 sheet hydrogen bonding interactions by two residues. Thus, the environment of the Trp222 and Trp245 side chain may be altered following conformational change, which results in quenching of Trp222 and (or) Trp245 fluorescence (Figure 22). Analysis of the intrinsic fluorescence time trace revealed that rapid fluorescence quenching was a combination of two different fluorescence decreases. The faster observed rate of fluorescence quenching (kQa) was -7-fold more rapid than the slower observed rate (kQb). Unlike fluorescence quenching, the subsequent recovery of fluorescence was monophasic, and exhibited an observed rate (kR) similar to kQb. Dependence of the observed rate constants on dsUA-25-mer DNA 154

concentration is shown in Figure 23G. The results showed that the rapid rate of

fluorescence quenching(kQa)(/1/led squares) was dependent on DNA concentration and reached the maximum rate (-150 s1) in the addition of 250 nM dsUA-25-mer.

The slow rate of fluorescence quenching(kQb)(closed triangles) and fluorescence recovery (kR, open circles) appeared to be independent of DNA concentration, and remained at -20 (Fig. 23 C).

4.1.4. Effect of NaGl onUNGFluorescence Quenching Associated with Binding Duplex Uracil-DNA

Previous reports suggested that disturbance of UNG-DNA electrostatic interactions influenced enzyme catalysis (70,128). To examine the effect of monovalent ions (NaC1) on the quenching of IJNG fluorescence associated with uracil-DNA binding,UNGwas reacted with dsUA-25-mer DNA in the presence of 0- 150 mM NaC1, and the change inLJNGfluorescence was monitored. As shown in Figure 25A, the magnitude of fluorescence quenching and fluorescence recovery were gradually reduced in the presence of NaC1. The reduction in fluorescence quenching correlated with the decrease in kQa (Figure 25B, closed squares), but not with kci, or kR

(Figure 25B, closed triangles and open circles, respectively). BothkQband kp were insensitive to NaC1. ThekQawas increased by 34% in the presence of 25 mM of NaC1. This result was consistent with previous reports that low concentrations of NaC1 stimulated uracil-DNA glycosylase activity, whereas high concentrations were inhibitory (62,100,178). These results suggested that perturbation of UNG-DNA electrostatic interactions influenced both enzyme catalysis and the fast phase of enzyme conformational change in a similar manner. Thus, catalysis and conformational change were closely associated. 155

Figure 20. Fluorescence properties of 2AP-containing oligonucleotide. A, fluorescence excitation (dashed line) and emission (solid line) spectra of double- stranded dNJTJ2AP-25-mer in buffer C. Single-stranded 2AP-25-mer (400 nM) was annealed to single-stranded dU-25-mer (440 riM) in buffer C. The real-time fluorescence emission spectrum was recorded at Xem=37O rim (Aex=3lO rim) at 25°C. B, quenching of single-stranded 2AP-25-mer fluorescence upon mixing with single stranded dU-25-mer. Single-stranded 2AP-25-mer (400 nM) was combined with increasing amounts (0 (dashed line), 50, 100,200, 300,400, 500, 600 and 800 riM) of single-stranded dU-25-mer in buffer C (a-h, respectively). The real-time fluorescence emission spectrum for each addition was recorded at Xem37O nm (Xex3 10 nm) after mixing at 25°C for 10 mm. Arrows indicate the wavelength at which the maximum fluorescence emission occurred. 156

ru

120 370nm /\V Wavelength (nm)

0

C.) U)

350 400 450 500 Wavelength (nm)

Figure 20 157

Figure 21. DNA binding affinity of UNG and R276X mutants for dU'2AP-25- mer. Changes in the 2AP fluorescence intensity of diiiU.2AP-25-mer at 370 tim were determined as a function of UNG (A) or R276C (B) concentration. dNJU2AP-25-mer (100 riM) was combined with 0-2 jiM UNG (A) or R276C mutant protein (B) in buffer C (2 ml), and the total fluorescence intensity at each addition (open squares) was monitored at Xem370 tim (2ex=31O nm) after mixing at 25°C for 5 mm. The fluorescence intensity of each protein concentration (open triangles) without DNA substrate was also recorded under the same conditions. The net fluorescence intensity (closed circles) at each protein concentration was obtained by subtracting the protein fluorescence intensity from the total fluorescence intensity. The net fluorescence intensities were a best fit to Equation 1 as described inSection 2.2.18,and the obtained Li values for UNG and R276C were 185 ±7 riM and 490±49 nM, respectively. The Li values for the remaining R276X mutant proteins were determined in similar fashion and are reported in the text. C, the binding affinity of each R276 mutant protein, indicated by the corresponding single letter amino acid abbreviation, for dijñJ2AP-25-mer was obtained by calculating 1/Kj for each protein and graphed relative to wild-type UNG (WT). 158

A. d

'nI-

I0 L

500 100015002000 UNG (nM) B. 10

j<2

500 100015002000 R276C (nM) c 0.06

0.04

0.02 WTCEH LWY Enzyme

Figure 21 159

Figure 22. A global conformational change of UNG* on binding iyU-DNA. Superposition of free UNG* (green) (111) and DNA-bound LTNG* (dark blue) in complex with uncleaved qiU-DNA (red) (117) shows that T.JNG* undergoes an conformational closing upon binding substrate DNA. The closure of a central gap between1 and 33 extended hydrogen bonding interactions by allowing inter--sfrand H-bonds, and was termed a J3-zipper (113,114). Closing of the J3-zipper effectively clamped the enzyme around the flipped-out uracil nucleotide and created the catalytically competent enzyme active site by bringing L272 into the DNA base stack, H268 and D145 into the active center, and inserting the leucine-intercalation loop into the DNA minor groove to hinder return of the uracil nucleotide to the base stack (117). 160

j3 1E1

Figure 22 161

Figure 23. Stopped-flow time trace of UNG intrinsic protein fluorescence upon binding double-stranded DNA. The intrinsic protein fluorescence of UNG (100 nM) was monitored in a stopped-flow spectrometer after mixing with 1 j.tM dsTA-25-mer (A) or dsUA-25-mer (B). Each trace representsan average often individual

acquisitions. The solid lines represent the best-fit curves to Equation 2. B, toppanel: residual analysis of each trace to the fit curve. C, the intrinsic protein fluorescence change of UNG (100 riM) was monitored in a stopped-flow spectrometer after mixing with 0-2 .tM dsUA-25-mer, respectively. The observed rate constants of individual kinetic trace were obtained by fitting each trace to Equation 2. The filled squares and filled triangles represent the rates of rapid (kQa) and slow (kQb) fluorescence quenching, respectively. The open circles represent the rate of slow fluorescence recovery (kR). The solid line represents the best-fit curve to the following equation:

k9a x [DNA kQa = kQa (Equation 5) Kd + [DNA] 162

A.

22

2.1

2.0

1.9 U- 0.05 0.10 0.15 Time (Seconds)

0.1 0.0 01 Ii U w

0.05 0.10 0.15 Time (Seconds) C.

200

150

j100

50

500 1000 1500 2000 LJA-25-mer (nM)

Figure 23 163

4.1.5. Effect ofArg276 Mutations on UNG Fluorescence Change Associated with Binding Duplex Uracil-DNA

To examine the effect of Arg276 mutations on the change in UNG intrinsic fluorescence associated with binding uracil-DNA, each R276 mutant protein was mixed with dsUA-25-mer DNA and the change in protein fluorescence was monitored. As shown in Figure 26, the change in intrinsic fluorescence of each mutant protein differed from that of wild-type UNG in several respects. First, the extent of mutant protein fluorescence quenching was significantly reduced relative to UNG (compare to Figure 23B). Second, the fluorescence recovery phase was not detected for any mutant protein during the 250 ms time scale. A longer recovery period (2 to 4 seconds) was required before intrinsic fluorescence returned to initial intensity levels. One exception was the fluorescence of R276E, which remained essentially unchanged (Figure 26, R276E). Third, the magnitude of fluorescence quenching of R276C, R276H, R276L, R276W, and R276Y mutant proteins was reduced, and resembled that of UNG at various concentrations of NaC1 (Figure 25). Data analysis revealed that the kQa of each R276 mutant protein was significantly reduced by 28 to 98 % compared to UNG (Figure 28A, solid black bars); the kQa of R276E had the greatest reduction. However, both the kQb andkRvalues of each mutant were essentially the same as those of UNG (Figure 28A, open and hatched bars, respectively).

4.1.6. Pre-steady State Kinetic Analysisofthe UNG Fluorescence Change Associated with Binding Single-stranded Uracil-DNA

In Chapter Three, the mutational effects of substitution at Arg276 on UV- crosslinking to PU-DNA were less severe in single-stranded DNA compared to double-stranded DNA (Figure 17). In addition, UNG was reported to remove uracil more rapidly from single-stranded (ss) DNA than from double-stranded (ds) DNA (100). Therefore, to further investigate the effect of Arg276 mutations on the change in UNG intrinsic fluorescence upon binding ssDNA, UNG was mixed with ssU-25- 164

mer and monitored for changes in intrinsic fluorescence. As shown in Figure 27, UNG (WT) exhibited rapid quenching of fluorescence in the first 15 milliseconds, followed by a slow fluorescence recovery over -45O ms. However, the fluorescence recovery from binding single-stranded U-DNA was faster than that observed for double- stranded U-DNA. Overall, the ssU-25-mer kinetic trace of UNG fluorescence quenching resembled that of UNG with dsUA-25-mer (Figure 27, WT, compared to Figure 23B). Moreover, fluorescence quenching was specific for uracil-DNA, since no fluorescence change was observed when UNG reacted with ssT-25-mer DNA. Next, the R276C, R276E, R276H, R276L, R276W, and R276Y mutants were reacted with ssU-25-mer and monitored for intrinsic tryptophan fluorescent change. Analysis of the stopped-flow traces showed that the intrinsic fluorescence change of each mutant protein was analogous to that of wild-type UNG (Figure 27, WT), exhibiting rapid fluorescence quenching in the first 15 milliseconds, followed by slower fluorescence recovery over 15O milliseconds (Figure 27). However, the recovery of R276B fluorescence to the starting level required more time (O.2O.3 s) than did UNG. Data analysis revealed that the kQa of each R276 mutant protein was essentially the same as that of UNG (Figure 28A, solid black bars), except that the kQa of R276E was reduced by -6O %. Both the kQb and kR of each mutant were very similar to those of UNG (Figure 28A, open and hatched bars, respectively).

4.1.7. Effect of Mutations at Arg2 76 on Uracil-Excision Activity

Since the intrinsic protein fluorescence change correlated well with enzyme catalysis and the effect of mutations at Arg276 on intrinsic protein fluorescence were minimal in reactions containing single-stranded uracil-DNA (Figure 27), it was of interest to learn how these mutations would affect catalytic activity on single- versus double-stranded uracil-containing DNA. The uracil base excision activity of UNG and the R276 mutant proteins was determined in reactions utilizing 5'-end labeled carboxyfluorescein (PAM) dsU'A-25-mer or ssU-25-mer DNA substrate, and the 165

reaction products were resolved by denaturing polyacrylaniide gel electrophoresis and quantified with a Hitachi scanning fluorometer (FMBioII). Control reactions showed that >95% of the 5'-FAM-dsU'A-25-mer substrate was converted to 5'-FAM-lO-mer product byE. coli Ung (Figure 29A, lane C). The minor amount (<10%) of single- stranded 5'-FAM-ssU-25-mer DNA substrate that was apparently refractory to cleavage by E. coli Ung was also refractory to cleavage by UNG (Figure 29B, lanes C and 1, respectively) and did not materially impact the experimental results. Titration of the 5'-FAM-dsUA-25-mer cleavage reaction with increasing amounts of UNG (Figure 29A, lanes 10-1) showed that approximately 50 % of the substrate DNA was converted to product in reactions containing .-40 Imol of UNG (Figure 29A, lane 8). In contrast, when UNG was replaced by equal molar amounts of R276E and assayed using the 5 '-FAM-dsUA-25-mer substrate under the identical reaction conditions, little 5'-FAM-lO-mer product was observed at any R276E amount less than 1.6 pmol (Figure 29A, lanes 12-20). When UNG was reacted with the single-stranded 5'-FAM- ssU-25-mer DNA substrate, approximately 15 fmol of enzyme resulted in 50 % product (Figure 29B, lanes 7). When the single-stranded DNA uracil excision assay was repeated with R276E (Figure 29B, lanes 11-20), the results were essentially indistinguishable from those obtained with wild-type UNG; that is, R276E was as active on single-stranded uracil-containing DNA as UNG. The uracil base excision assay was repeated for R276C, R276H, R276L, R276W, and R276Y. UNG was assayed with each R276 mutant enzyme preparation, and the reaction products were analyzed by denaturing polyacrylamide gel electrophoresis as described. Quantification of the substrate (S) and product (P) bands in each gel was carried out, and the R276 mutant enzyme activity was compared to that of UNG, which was set to 100 %. As shown in Figure 29C, mutations at Arg276 significantly reduced uracil base excision activity on double-stranded DNA, as R276C, R276E, R276H, R276L, R276W, and R276Y retained 31, 0.5, 35, 33, 30, and 25 %, respectively, of UNG activity. In sharp contrast, the activity of the R276 mutant enzymes on ssU-25-mer was equivalent to UNG, since R276C, R276E, R276H, R276L, R276W, and R276Y exhibited 94, 89, 99, 93, 97, and 100 % of UNG activity (Fig. 29D).

4.2. Discussion

In 2AP fluorescence measurements under steady-state conditions, the UNG concentration-dependent 2AP fluorescence enhancement did not appear to be entirely linear at lower enzyme concentration range (< 100 nM) (Figures 20A). This may be due to the slight excess (approximately 10 %) of single-stranded dU-25-mer present in the substrate preparation after annealing to the complementary single-stranded 2AP- 25-mer. Previous studies indicated that UNG was more active (-'3-fold) on uracil- containing single-stranded DNA than on double-stranded DNA (100). Therefore, it was possible that a small percentage of UNG in the 2AP fluorescence measurements bound preferentially to the non-fluorescent single-stranded dU-25-mer, reducing the total 2AP fluorescence monitored by the spectrofluorometer and perturbing trueJ(D value determination. However, since the same DNA substrate preparation was used in binding measurements of both UNG and the R276 mutant proteins, determination of the relative DNA binding affinities for dU'2AP-25-mer should not have been affected. Jiang and Stivers (126) studied the effect of replacing the key leucine loop residue Leul 91 in E. coli Ung with alanine. The extrahelical state of the uracil nucleoside analog 2'-fl-fluoro-deoxyuridine opposite adenine was assessed by 2AP- fluorescence, and the enzyme conformational change by initial quenching of intrinsic protein fluorescence (126). Their data indicated that flipping of the uracil nucleotide analog by the L191 mutant protein was unimpaired; however, quenching of the intrinsic protein fluorescence that typically occurs concomitantly with base flipping was not observed (126). These results suggested that the LI91A mutant protein did not effectively isomerize to the closed conformation required to lock the uracil residue in the active site (126). The catalytic activity of the analogous UNG mutation, L272A, 167

was less than 1 % of wild-type(116).Stiverset al.(118) hypothesized that the change in uracil-DNA glycosylase tryptophan fluorescence most likely reflected the global enzyme conformational change associated with the leucine-loop insertion step. As

demonstrated in this Dissertation (Figure23B),the fact that the change in UNG tryptophan fluorescence was independent of uracil-DNA concentration suggested that UNG followed a multi-step binding mechanism similar to that proposed for E. coil Ung (118): 1) formation of a concentration-dependent weak nonspecific enzyme-DNA complex, followed by 2) reversible concentration-independent uracil-flipping, and 3)

conformational change around the flipped-out base(126).

As suggested by Wonget al. (121),penetration of the leucine-loop into the DNA minor groove appears to act more as a "doorstop" to prevent the return of the flipped-out uracil residue than as a piston to push the uracil nucleotide out of the DNA helix. This hypothesis is consistent with the observation by Jianget al.(303) that the E. coli Ll91 mutation could be rescued if a pyrene nucleotide were incorporated in the DNA strand opposite to the uracil base. The bulky pyrene moiety was thought to maintain the uracil nucleotide in an extrahelical state, since it filled most or all of the space in the DNA helix normally occupied by a dsU'A base pair. Extrapolating from their results, Jianget al.(303) described the role of the Leul91side chain as a "plug" to tilt the equilibrium of uracil flipping toward the enzyme active site pocket rather than the helical base stack. Based on the structural and kinetic work discussed above, the reduced kQa exhibited byR276mutant proteins for dsUA-25-mer DNA was caused by defective leucine-loop intercalation which allowed the flipped uracil nucleotide to return to the base stack. Thus, theR276mutant proteins were less efficient at capturing the uracil base into the uracil-specificity pocket, and the rate of enzyme conformational change was reduced. Failure to lock in the flipped uracil and generate the active site through conformational change led to greatly reduced uracil- excision activity (Figure29B). Several results suggested that the fast rate of fluorescence quenching, kQa, might indicate docking of the uracil nucleotide in the UNG active site pocket; W245 is well positioned to report on the initial conformational change brought about by

docking (see Figure 24). First, the influence of NaC1 concentration on UNG kQa correlated with the effect of NaC1 on UNG catalytic activity. The inhibitory effect of increasing NaCI concentration on UNG catalytic activity is generally thought to be electrostatic: the affinity of the positively-charged enzyme DNA binding site for the

polyanionic DNA molecule is decreased by ion shielding. Second, the kQa of TJNG and the R276 mutant proteins correlated well with their equilibrium DNA binding

affinities. Thus, the similar kçavalues of UNG and the R276 mutant proteins (except R276E) for ssU-25-mer suggested that the binding affinity of these proteins for 5sDNA were similar and that the Arg276 guanidinium side chain was dispensable for uracil flipping and enzyme conformational change when the target base was located in ssDNA. Notably, the performance of the least catalytically active mutant, R276E, was greatly improved on the ssDNA (Figure 29C). Parikh etal. (116) deduced from the UNG*U.G DNA co-crystal structure that the EN of Arg276 formed a water-mediated hydrogen bond with the Leu272 carbonyl and the N3 of the adenine (Ade6) nucleotide 3' to the uracil nucleotide. These water-mediated interactions might not be necessary for the Leu272 side chain to facilitate docking of the uracil-nucleotide in the enzyme active site pocket. Uracil flipping might be easier in ssDNA because base stacking and hydrogen bonding interactions are greatly reduced in ssDNA, and the sugar-phosphate backbone is more flexible than in dsDNA. The uracil nucleotide in ssDNA can be flipped without generating torsional strain in the phosphate backbone; therefore, it has a better chance of docking in the uracil-specificity pocket, where it can make the high

affinity interactions that result in enzyme conformational change. The fact that the kQa of UNG and R276 mutant proteins was higher for ssU-25-mer than for dsUA-25-mer suggested that the binding affinity of UNG and the R276 mutant proteins for ssDNA was greater than for dsDNA. Thus, mutations at Arg276 efficiently transform UNG into a single-strand-specific uracil-DNA glycosylase. The slower rate of UNG fluorescence quenching, kQb, and the slow rate of fluorescence recovery,kR,may represent isomerization of the enzyme from the open 169

to the closed, or from the closed to the open conformation, respectively. It is interesting to speculate that the fluorescence quenching of W222 may reflect this isomerization step, since W222 is farther from the active site than W245, and the rate of its fluorescence change is slower. The magnitude of kQb and kR are almost identical, and both kQb and kR are independent of uracil-DNA concentration. The kQb and kR for ssDNA are almost the same as those for dsDNA. In addition, both kQl, and kR are insensitive to NaCl andMgC12(See next Chapter). These results suggested that mutations at Arg276 did not affect the rate of enzyme conformational change once the uracil nucleotide was productively bound. 170

Figure 24. Tryptophan residues of human uracil-DNA glycosylase. A, overlay of the Ca backbone structure of human uracil-DNA glycosylase free and bound to DNA. The tertiary structures (116) of UNG* free (blue) and UNG* L272A in complex with abasic site-containing DNA (DNA not shown) were overlayed by Insightil and displayed a root mean square of 1.4A.The eight tryptophan residues are displayed for UNG* free (white) and UNG* L272A (green). The amino- (N) and carboxy-terminus (C) are labeled for both proteins (302). B, the illustration of W222 (orange) and W245 (yellow) of UNG*. a-Helices are depicted as green cylinders and 3-sheets are illustrated as blue strands. Deoxypseudouridine (NJU) was marked as red. Structures were drawn with the Cn3D 4.0 software program using the PDB file 1EMH (MIIVIDB: 13471) deposited by Parikh et al. (117) in the Molecular Modeling Database of the National Center for Biotechnology Information. 171

A.

Figure 24 172

Figure 25. Effect of NaCl concentration on dsU.A-25-mer-induced UNG intrinsic fluorescence. A, UNG (100 nM) was mixed at 25 °C with 1 tM dsUA-25-mer containing 0, 25, 50, 75, 100, or 150 mM of NaC1 in buffer C as indicated and the fluorescence intensity was monitored in a stopped-flow spectrometer. Each trace shown represents an average of 10 individual acquisitions. The solid lines represent the best-fit curves to Equation 2. B, the observed rate constants for fluorescence quenching (kQa,fihled squares and kQa,fihled triangles) and fluorescence recovery(kR, open circles) were obtained by fitting each kinetic trace to Equation 2 as described in Section 2.2.20 and graphed as a function of NaC1 concentration. 173

2.4 +0mM NaC 2.0'

2.4 +25 mM NaCI

2.0

2.4 +50 mM NaCJ >0 0 U 2.0

U 2.4 +75 mM NaCI 0) 0

U-

2.4 +100 mM NaCI

2.4 +150 mM NaCI

2.0

0.05 0.15 0.25 Time (Seconds) B. 250

. 200

150 .0 0 100

50

I

NaCI (mM)

Figure 25 174

Figure 26. Stopped-flow time trace of R276 mutant protein intrinsic fluorescence change upon binding dsUA-25-mer. Each R276 mutant protein (100 nM) was mixed with dsUA-25-mer (1 j.iM) in buffer C at 25 °C and the intrinsic protein fluorescence monitored using a stopped-flow spectrometer. Each kinetic trace shown represents an average of 10 individual acquisitions; the identity of the R276 mutant protein is indicated in the upper right corner of the time trace. The solid lines represent the best-fit curves to Equation 2 as described inSection 2.2.20. 175

R276C

2.3

21

R276E 2.3

2.1

4I R276H 2.3

2.1 a, R276L U) a 2.3 0

LI.. 2.1

R276W 2.3

2.1

R276Y 2.3

2.1

0.05 0.15 0.25 Time (Seconds)

Figure 26 176

Figure 27. Effect of single-stranded U-25-mer DNA on the intrinsic protein fluorescence of UNG and R276 mutant proteins. Each R276 mutant protein (100 nM) was mixed with ssU-25-mer DNA (1 jtM) in buffer C at 25 °C and the intrinsic protein fluorescence monitored using a stopped-flow spectrometer. Each kinetic trace shown represents an average of 10 individual acquisitions; the identity of the wild- type UNG and R276 mutant proteins is indicated in the lower right corner of the time trace. The solid lines represent the best-fit curves to Equation 2 as described inSection 2.2.20. 177

IA

I .

1.2 1.4

1.3

R276C I 1.2 1.4

1.3 (I)

o 1.2 R276EJ

0 C) 1.3 0 C) R276H1 1.2 0- I

U 1.3

R276L I 1.2 1.4

1.3

1.2

1.3

1.2 R276Y1

0.05 0.10 0.15 Time (Seconds)

Figure 27 178

Figure 28. Observed rate constants for ssU-25-mer- and dsUA-25-mer-induced intrinsic protein fluorescence change. A, the observed rate constants associated with dsU'A-25-mer were obtained from Figures 23B and 26. B, the observed rate constants associated with ssU-25-mer were obtained from Figtire 27. The filled squares and filled friangles represent the rates of rapid (kQa) and slow (kQb) fluorescence quenching, respectively. The open circles represent the rate of slow fluorescence recovery (kR).The R276 mutant proteins are identified by single-letter amino acid abbreviations; WT stands for UNG. 179

A.

0

WI C E H L W Y Mutant Enzyme B. ssU-25-mer

.0 0

Mutant Enzyme

Figure 28 Figure 29. Uracil base excision activity of IJNG and R276X mutant proteins on single- or double-stranded uracil-containing DNA. UNG and R276X mutant proteins (fraction N) were assayed for uracil-DNA glycosylase activity using the 5'- end carboxyfluorescein (FAM)-labeled double- and single-stranded uracil-containing oligonucleotide substrates, 5'-FAM-dsUA-25-mer (A) and 5'-FAM-ssU-25-mer (B). Control reactions consisted of: S, substrate alone; M, mock reaction mixtures containing substrate in reaction buffer but lacking enzyme; and C, reaction mixtures to which E. coli Ung (88 finol) was added (lanes S, M, and C, respectively). Reactions contained 1.6 pmol, 80, 40, 30, 20, 17.5, 15, 10, 8.75, and 4.375 fmol of UNG or

R276E, as indicated, inlanes1-10, and 11-20, respectively. Following incubation at 30 °C for 15mm,reaction mixtures that contained duplex 5'FAM-dsUA-25-mer (A), were subjected toEndo 1V(1 unit) treatment to cleave abasic (AP) sites generated by uracil excision, while the reaction mixtures containing the single-stranded substrate 5'FAM-ssU-25-mer (B) were subjected to hot alkaline treatment in order to cleave AP sites. Reaction products were analyzed by 12 % denaturing polyacrylamide gel electrophoresis and the gels were scanned using a FMBioII fluorescence imaging system. Arrows indicate the locations of the 5'-FAM-ssU-25-mer oligonucleotide substrate (3) and 5 '-FAM- 1 0-mer uracil-excision product (P). Results were analyzed using the ImageQuant program. The uracil-DNA glycosylase activity of the R276 mutant proteins toward double- or single-stranded uracil-containing DNA (C & D, respectively) was normalized to that of UNG, which was defined as 100 %. R276X mutant proteins are denoted by single-letter amino acid abbreviations. Error bars represent the standard deviation of three experimental determinations. 181

A. dsDNA SM C ______

.4p

+f7E -SMCr-----_ - ----*---- .4p

C.

75

50 r25 UNG Mutant D.

UNG Mutant

Figure 29 182

5. Biochemical Characterization of the Nuclear Form Human Uradil-DNA Glycosylase: Effect of NaCI andMgCl2on DNA Binding, Uradil Flipping, and Enzyme Conformational Change

In previous two chapters, all experiments were performed with UNG*, UNG and UNG mutant enzymes. However, UNG* and UNG are artificial enzymes created in the laboratory, and do not exist as real enzymes in human cells. As mentioned in "iNTRODUCTION", UNG* and UNG are N-terminal truncation mutants and contain only the catalytic domain common to both UNG1 and UNG2. The unique N-terminus of UNG1 and UNG2 appears to determine the role each enzyme will play in the cell. For example, the N-terminus of UNG2 contains RPA and PCNA binding motifs, whereas the N-terminus ofUNG1 encodes a mitochondrial localization sequence (105). Hence, it is important to know how these N-terminal domains affect or modulate the properties of the core catalytic domain of the enzyme. This chapter presents the results and interpretations of biochemical analysis of UNG2 with regard to the influence of NaC1 andMgC12on DNA binding, uracil flipping, conformational change, and catalytic activity. Recombinant UNG2 was purified to apparent homogeneity by conventional chromatography. The specific activity of the purified enzyme was then determined relative to UNG and its ability to bind to Ugi evaluated. Transient kinetics was employed to study the effects of NaCl andMgCl2on uracil flipping and UNG2 conformational change. The rates of UNG and UNG2 uracil flipping and enzyme conformational change were compared. The results suggest that the N-terminus of TJNG2 plays an important role in the enzyme's response toMg2.

5.1. Results

5.1.1. Overproduction and Pur/Ication of Recombinant UNG2

The full-length UNG2 protein has been notoriously difficult to produce and purify in many biological systems, mainly due to in vivo proteolytic attack as well as 183

degradation of the UNG2 amino terminus during purification. Recently, Kavli etal. (149) reported that the N-terminal degradation problem could be partially overcome by expressing the protein with an N-terminal His-tag. Thus, in order to avoid N-terminal degradation and ease purification, the UNG2 sequence was cloned into a pET28a vector to produce a His-tagged UNG2 fusion protein. UNG2-pBT28a vector was then transformed into E. coli strain BLR supplemented with pRIL, and purified as described in Section 2.2.22. For functional analysis, the His-tag was removed from the fusion protein by treatment with thrombin in the final stage of purification (Section 2.2.22). Cleavage of the nineteen amino acid His-tag by thrombin left behind three amino acids, Gly-Ser-His, on the N-terminus of UNG2. To examine the purity and relative gel mobility of each fraction (fraction 1VI) during UNG2 purification, 2 jig of each fraction was subjected to 12.5% SDS- polyacrylamide gel electrophoresis. Analysis of the Coomassie-stained gel showed that UNG2 (fraction VI) was purified (> 98%) to apparent homogeneity (Figure 30A). To ascertain the integrity of the N-terminus of UNG2 after thrombin cleavage, UNG2 (fraction VI) was subjected to matrix-assisted laser desorption-ionization (MALDI) mass spectrometry as described in Section 2.2.23. As shown in Figure 30B, UNG2 appeared as a single symmetric peak, corresponding to Mr 35,435 Da, which was consistent the deduced molecular weight (-35,500 Da) of UNG2 plus three extra Gly- Ser-His amino acids. These results indicated that UNG2 (fraction VI) was apparently homogeneous and that the product of thrombin cleavage was as predicted.

5.1.2. Ability ofUNG2 toBind Ugi

In order to ascertain whether the purified UNG2 protein was properly folded, it was reacted with the uracil-DNA glycosylase inhibitor protein (lJgi), as described in Chapter Three. To scrutinize the structural integrity of recombinant UNG2, 40 pmol of UNG2 protein was reacted with a 2.5-fold molar excess of Ugi (100 pmol), and the reaction products were analyzed by 10 % non-denaturing polyacrylamide gel 184

electrophoresis. As shown in the Figure 30C, UNG2 did indeed form a complex with Ugi, as the appearance of the UNG2.Ugi complex band was concomitant with the disappearance of UNG2 band (lane 5). These results established that UNG2 protein was properly folded and the conserved DNA-binding pocket was not disturbed.

5.1.3. The Relative Uracil-DNA Glycosylase ActivitiesofUNG2, UNG* and UNG

To determine the relative specific activities of UNG2, UNG* and UNG, 0.04- 0.13 units of each protein was introduced to the standard reaction mixture containing 7.9 nmol of calf thymus {uracil-3H] DNA (175 cpmlpmol of uracil). The amount of [3H]uracil releasewas measured as described in Section 2.2.12.1. The specific activity of each protein (units/mg) was then calculated and the relative specific activity of UNG2 and UNG* was normalized to UNG activity, which was set as 100%. As shown in Figure 30D, the specific activity of 1JNG2 relative to UNG or UNG* was significantly reduced. It was about 11.5 % of UNG or 7 % of UNG* activity (Figure 30D, UNG2 bar). This finding is consistent with the previous reports (100,149) that addition of extra amino acid sequences to the N-terminus of UNG* resulted in reduction of enzyme activity.

5.1.4. NaC1 andMgC12 Effectson UNG2 Activity

Kavli and co-workers (149) previously showed that UNG2 activity was strongly stimulated by physiological concentrations of Mg2. To investigate the influence of NaC1 and MgC12 on UNG2 activity, one unit of UNG2 and 0.06 units of UNG or UNG* were introduced into standard reaction mixtures containing increasing concentrations of NaCI or MgC12 and the extent of uracil excision was determined (Figure 31). The activity of each enzyme was determined for the various treatment conditions, and normalized to the zero NaC1 orMgCl2reaction, which was set as a 100%. As shown in Figure 3 1A (filled circles), UNG2 activity was stimulated4-fold by 50 mM of NaC1 and declined gradually at NaC1 concentrations greater than 100 mlvi. The activity of UNG* and UNG declined gradually in the presence of NaC1 and was inhibited by -99 % at 200mM of NaC1 (Figure 31A,fihled squares and triangles, respectively). UNG2 activity was stimulated about 5-fold at 8 mM MgCl2, and slowly declined as the concentration of MgC12 increased (Figure 3 1B,fihled circles). In contrast, the activity of TJNG* (fIlled triangles) and UNG (fIlled squares) activity increased by less than 15 % at 4 mM of MgC12, and declined gradually as the MgCl2 concentration increased (Figure 3 1B,filled squares and triangles). These results indicate that the stimulatory effects of NaCl and MgCl2 on enzyme activity are unique to UNG2.

5.1.5. Effect of NaCI and MgCJ2 on UNG2-Binding to Duplex PU-DNA

To examine the effect of NaC1 and MgC12 on the DNA binding/base flipping activity of UNG2 under equilibrium conditions, a fluorescence technique described in Figures 16 and 21 was used. As shown in Figure 32A, the enhancement of 2AP fluorescence correlated with increasing UNG2 concentration, and saturated at 400 nM. This 2AP fluorescence enhancement was UNG2-dependent, since pre-incubation of UNG2 with Ugi prior to the reaction totally abolished 2AP fluorescence enhancement. The hyperbolic dependence of 2AP fluorescence on UNG2 concentration was fitted to Equation I as described in Section 2.2.18. The dissociation constant (Li) of UNG2 for binding to dNJU2AP-25-mer was determined to be 270±35 nM, which was higher than that of UNG (185 ± 7 nM) (Figure 21A). Therefore, UNG2 bound more weakly than UNG to this substrate. Next, the effect of NaC1 and MgCl2 on UNG2 equilibrium binding to iiU-DNA was examined. The fluorescence change of reaction mixtures containing 500 nM of UNG2 and 100 nM of duplex dU2AP-25-mer in the presence of NaC1 (0-150 mM) or MgCl2 (0-15 mM) was monitored. The results showed that NaC1 and MgCl2 diminished UNG2-induced 2AP fluorescence, which was totally abolished at 50 mM 186

NaC1 or 8 mMMgC12(Figure 32B & C, closed circles, respectively). A similar NaC1 effect on 2AP fluorescence quenching was observed for UNG (Figure 32B, open circles). However,MgC12(<1 mM) slightly enhanced (<5 %) the 2AP fluorescence by UNG (Figure 32C, open circles). These results demonstrated that NaC1 andMgC12 reduced the equilibrium DNA binding/base flipping activity of UNG2.

5.1.6. Pre-steady State Kinetics of UNG2-binding to dsU2AP-25-mer

Since iiiU cannot be not hydrolyzed by UNG (117), a uracil-containing DNA substrate was used to determine the base flipping activity induced by UNG2 binding. The fluorescence properties of dsU2AP-25-mer are shown in Figure 40. To examine the pre-steady state kinetics of UNG2 DNA binding and uracil flipping, stopped-flow experiments were performed using 200 nM of dsU2AP-25-mer and 1 tM of UNG2 as described in Section 2.2.19. As shown in Figure 33B, the 2AP fluorescence trace showed a rapid and transient enhancement within the first 10 ms. This 2AP fluorescence enhancement was UNG2-dependent, since the fluorescence change was not observed in the reaction without UNG2 addition (Figure 33A). The obtained stopped-flow trace was fitted to Equation 3 as described in Section 2.2.18, and the observed rate of 2AP fluorescence enhancement was determined to be 296 si. The dependence of the observed rate on TJNG2 concentration was obtained by titration of 200 nM of cjsU'2AP-25-mer with 400-4000 nM of UNG2; the rate constants ranged from 120 to 440 s_i (Figure 33B). The non-linear dependence of the observed rate constants on UNG2 concentrations ruled out a direct one-step uracil flipping mechanism concomitant with DNA binding.

5.1.7. Effect of NaC1 orMgC12on UNG2 DNA Binding and Uracilflipping

The effect of NaC1 andMgC12on UNG2 DNA binding and uracil flipping was monitored by stopped-flow experiments using 200 nM of dsU2AP-25-mer with 1 M 187

UNG2 in the presence of 0-150 mM of NaCI or 0-15 mM of MgC12. As shown in Figure 33D, increasing concentrations of NaCI caused a gradual reduction in the observed rates, and 100 mM NaC1 reduced the observed rate to near zero. Similar results were obtained with UNG2 and UNG in the presence of MgC12 (Figure 34, closed circles and open squares, respectively). MgC12 concentrations higher than 8 mM essentially abolished 2AP fluorescence change for UNG2 and UNG. These results were similar to those shown in Figure 32B & C. Therefore, high NaC1 (>100 mM) and MgC12 (>8 mM) concentrations perturbed UNG2 DNA binding/uracil flipping under both equilibrium and pre-steady state conditions.

5.1.8. Pre-steady State KineticsofUNG2 Intrinsic Fluorescence Change Induced by Binding to Uracil-DNA

As shown in Figure 23, the conformational change of UNG upon binding duplex uracil-DNA can be monitored by enzyme intrinsic protein fluorescence change. To determine whether UNG2 would display an analogous change of intrinsic fluorescence in real time, the transient intrinsic enzyme fluorescence was monitored as described in Section 2.2.19. As shown in Figure 35B, when 200 nM of UNG2 was mixed with 1 jtM of dsU'A-25-mer, the stopped-flow trace showed rapid fluorescence quenching in the first 20 milliseconds, followed by a slow fluorescence recovery over 2 seconds. This kinetic trace was similar to that of UNG (Figure 23B). Data analysis supported the hypothesis that the observed fluorescence quenching was a combination of two different kinetic phases. One had a -40-fold faster rate (kQa) than the other (k). Unlike fluorescence quenching, the subsequent fluorescence recovery was monophasic, and showed an observed rate (kR) equal to kQb. Like UNG, the UNG2 fluorescence quenching was specific to binding uracil-DNA, since no fluorescence change was observed when UNG2 reacted with normal dsTA-25-mer (Figure 35A). The dependence of the rate constants observed for UNG2 on dsUA-25-mer DNA concentration was investigated and is shown in Figure 35G. The rates of fluorescence

quenching (kQa and kQb,filled squares and filled triangles, respectively) and 188

fluorescence recovery (kR, open circles) appeared to be independent of DNA

concentration, and were determined to be 220s1,18 si, and 18 si, respectively. These results were similar to those obtained rates with UNG (Figure 23 C).

5.1.9. Effect of NaC1 and MgC12 on UNG2 Intrinsic Fluorescence Change upon Binding Duplex Uracil-DNA.

To examine the effect of NaC1 concentration on the change in UING2 intrinsic fluorescence upon binding duplex uracil-DNA, stopped-flow experiments were performed as described in Section 2.2.19, except that 0-150 mM of NaC1 was incubated with 1 p.M of dsUA-25-mer prior to reaction with 200 nM of UNG2. As shown in Figure 36A, the magnitude of both fluorescence quenching and recovery were gradually reduced in the presence of increasing amounts of NaC1 (Figure 36A, top to bottom panels). Data analysis revealed that kca was enhanced by 45 % at 25 mM of NaCI (Figure 36B,fihled squares) and decreased as NaC1 concentrations increased. Both kQb and kR were not affected by NaCl (Figure 36B, closed triangles and open circles, respectively). Identical results were obtained when NaC1 was pre- mixed with UNG2. The effect of MgC12 on UNG2 and UNG intrinsic protein fluorescence change was further examined as described above, except that NaCl was replaced by 0-32 mM of MgCl2. As shown in Figure 37 (top to bottom panels), the amplititude of both UNG2 fluorescence quenching and recovery were gradually reduced in the presence of increasing amounts of MgC12. Similar results were obtained for UNG. Data analysis revealed that the kQa of UNG2 was increased by 94 % at 1 mM of MgCl2 concentration and declined when the concentration of MgCl2 increased (Figure 38A).

At 16 mM of MgC12, the kQa was reduced by 90 % (Figure 38A). In contrast, the kQa of UNG was increased by oniy 5 % at 1 mM of MgC12 concentration and declined

when the concentration of MgCl2 increased (Figure 38B). Similarly, the kQa of UNG was reduced by 90 % at 16 mM of MgC12 concentration (Figure 38A). However, both kQb and kR of UNG2 or UNG were essentially unchanged and remained at 17 s_i for 189

UNG2 and 20 sfor UNG (Figure 38A & B). Thus, MgC12 specifically enhanced the fast rate of UNG2 fluorescence quenching associated with binding duplex uracil- DNA.

5.1.10. Direct Comparison of Time Courses of2AP Fluorescence Enhancement and Intrinsic Protein Fluorescence Change Using a Duplex Uracil-DNA Substrate

Previously, Wong et al. (121) demonstrated that when E. co/i Ung bound to a duplex uracil-containing DNA, the resulting 2AP fluorescence enhancement occurred earlier than the change in protein fluorescence. The authors interpreted this result to mean that uracil nucleotide flipping occurred earlier than Ung conformational change (121). Since the experimental approach followed by Wong et al. (121) could be applied to the UNG2 protein, it was possible to examine whether uracil nucleotide flipping took place before the UNG2 conformational change. Stopped-flow experiments were performed with either 100 nM of dsU2AP-25-mer and lp.M of IJNG2 or 1pM of dsUA-25-mer and 100 nM of IJNG2. Direct comparison of the two stopped-flow traces was obtained at equal molar concentration of DNA or protein and their respective pseudo-first order components. As shown in Figure 39A, 2AP fluorescence enhancement (top panel) induced by UNG2 occurred at nearly the same rate as UNG2 fluorescence quenching (bottom panel). Detailed analysis revealed that the rate of 2AP fluorescence enhancement (kflp = 368 1)was faster than those of

UNG2 fluorescence quenching(kQa =228 s'). These data are consistent with those of Wong et al. (121) and indicate that, for dsU2AP-25-mer substrate DNA, uracil nucleotide flipping by UNG2 occurred before conformational change. Therefore, these results support the premise that UNG2 utilizes a "pinch-pull-push" mechanism. Since UNG2 bound duplex dipU2AP-containing DNA and induced base flipping (Figure 33), it was of interest to learn whether nucleotide flipping preceded protein fluorescence change with duplex dU-DNA. As shown in Figure 39B (top panel), the 2AP fluorescence enhancement induced by UNG2 was similar to that 190

shown in Figure 39A; however, the rate of fluorescence enhancement (kflp = -304s1) was slower than that observed for dsU'2AP-25-mer(kfl1p =-'368 s"). In contrast, when UNG2 bound dijiUA-25-mer, UNG2 intrinsic fluorescence quenching was not observed (Figure 39B,bottom panel).Therefore, binding of dijiU'A-25-mer to UNG2 was not sufficient to induce the UNG2 conformational change.

5.2. Discussion

Analysis of the co-crystal structure of UNG* in complex with Ugi revealed that Ugi binds the sequence-conserved DNA-binding groove of UNG* via shape and electrostatic complementarity, specific charged hydrogen bonds, and hydrophobic packing against Leu272, which protruded from the UNG* leucine intercalation loop (183). Thus, the ability of UNG2 to stably bind to Ugi not only indicated the DNA- binding pocket of UNG2 was properly folded, but also suggested that the presence of the additional N-terminal sequence inherent to UNG2 did not interfere with the formation of enzyme-inhibitor complex. Therefore, the location of the NH2-terminus of UNG2 was likely to be structurally removed from the UNG2Ugi interaction face and, hence, did not block the TJNG2DNA-binding pocket. This assumption is consistent with the previous structural model for the RPA32C-TJNG2-DNA complex, which suggested that the NH2-terminal peptide sequence from Arg77 to Arg88 of UNG2 was located opposite the DNA binding site (259). The observation in this Dissertation that the stimulatory effect of NaCl and MgCl2 on steady-state enzyme catalytic activity was specific for UNG2, but not

UNG* (or UNG), was in agreement with the report by Kavliet al.(149) that the unique NH2-terminal amino acids of TJNG2 were required for MgC12 or NaC1- stimulation. In contrast, the effect of MgCl2 on UNG2 in equilibrium and pre-steady state uracil-DNA binding experiments was inhibitory (Figures 32C and 34). This result was consistent with a previous study by Kavliet al.(149) that MgC12 did not increase the UNG2 DNA binding affinity for dsUA-DNA. In their experiments, 191

MgCl2 (7.5 mM) was found to increase the catalytic efficiency (kcat/Km) of UNG2 on

dsUA-DNA3.5-fold by increasingkeat 3.2-fo1d and reducing Km --3 %. However, when UNG2 was reacted with single-stranded uracil-containing DNA,MgC12 increased the kcat/Km quotient by reducing Km by 138-fold (149). Thus,MgC12had a more profound effect on steady-state UNG2 DNA binding affinity when the enzyme reacted with ssU-DNA. It is possible that MgCl2 may interact with either 1) the N- terminal domain of UNG2, or 2) with the single-stranded uracil-DNA substrate, in order to facilitate UNG2-DNA interactions. Coordination to Mg2 may shift the conformation of the N-terminus from a disordered state to a more organized conformation that interacts synergistically with the core catalytic domain to increase uracil-DNA binding affinity. The observation that UNG2 could form a stable complex with Ugi indicated that the ninety-three amino acid N-terminus of UNG2 was probably not proximal to the active site, since Ugi binds specifically to the DNA binding groove and leucine intercalation loop. Mer and co-workers (259) demonstrated that the unique NH2- terminal amino acids from Arg73 to Arg88 of UNG2(UNG7388)had no stable structure, but adopted a helical conformation upon binding the NH2-terminus of RPA32 (RPA32170270). Hence, it is possible that Mg2 causes a conformational shift in the UNG2 N-terminus. Reorganization of the N-terminus may promote intramolecular interactions that facilitate closing of the "3-zipper." Indeed, the fast rate of UNG2 intrinsic fluorescence quenching (kQa) associated with binding duplex uracil-DNA was enhanced by -94% in the presence of 1 mM of MgC12 (Figure 38A). On the other hand, Mg2 may bind to the single-stranded uracil-DNA substrate, affecting a conformation change in the DNA that renders the uracil nucleotide more accessible to nucleotide flipping. DNA interaction with Mg2 may act to coordinate the electrostatic affinity between the DNA binding grove of UNG2 and the uracil-DNA, leading to a reduction in Km. It should be noted that the higher concentrations of MgC12 (> 8 mlvi) are inhibitory. This negative effect may result from shielding of the negative charge of DNA phosphate backbone, which would reduce the UNG2-DNA electrostatic 192

interactions. Taken together, the data show that UNG2 contains aMg2response element that influences the catalytic activity of the enzyme. The TJNG2 rate of slow fluorescence quenching (kQb) and the rate of fluorescence recovery (kR) were not affected byMgC12(Figure 38A). As posited in the previous chapter, kQb and kR represent the forward and reverse rates of the UNG2 conformational isomerization. Interestingly,MgC12had no effect on kQb and kR when UNG2 bound duplex uracil-DNA. Similarly, the presence of NaC1 did not perturb the kQb and kR of UNG2 (Figure 36B). However, the observation that NaCl increased the

kQaof UNG2 was perplexing, since NaC1 also increased the UNG2 but not UNG catalytic activity (Figures 32A). Direct comparison of stopped-flow time traces of 2AP fluorescence enhancement induced by UNG2 and UNG2 intrinsic protein fluorescence change associated with binding uracil-DNA (Figure 39) showed that the rate of 2AP fluorescence enhancement (kflp) was faster than the rates of protein fluorescence quenching (kQaand kQb). This result was consistent with previous observations forE. co/i Ung (121) and suggested that uracil flipping occurred prior to UNG2 conformational change. Hence, uracil flipping ahead of enzyme conformational change may be common to the mechanism of family-i UDGs. In the UNG*.NJUDNA co-crystal structure, binding to the UNG* uracil- specificity pocket rotated the uracil ring by 9Ø0on its N1-C4 axis from its normal anti-conformation to a position halfway between anti and syn configuration (117). Since the ijiU base was still connected to the deoxyribose in the enzyme-DNA complex, the NJU deoxyribose was pulled0.4Adeeper into the enzyme active site pocket than in the enzyme-product complex (116). However, the 'qiU was sterically prevented from inserting as deeply into the active site center as the cleaved uracil in the enzyme-product complex. Therefore, the stacking interaction between i1iU base and Phe 158 aromatic side chain was less than ideal. Analogously, the hydrogen bonding between ijiU 02 and 04, and UNG* H268 and N204 side chains, respectively, would be weaker than that observed for uracil, as deduced from the 193 enzyme-product complex. Therefore, the reduced positive interactions between the U base and the enzyme active site residues were insufficient to trigger the enzyme conformational change, and no fluorescence change was observed. 194

Figure 30. Purity of purified recombinant UNG2. A, SDS-polyacrylamide gel analysis of recombinant UNG2 isolated at various step during the purification. Protein samples from the purification of recombinant UNG2 (fraction I-VT) containing 5,4, 3, 3,2, and 2 jig of protein, respectively, were analyzed by 12.5% SDS-polyacrylamide gel electrophoresis and stained with Coomassie Brilliant Blue G-250. The location of the protein molecular weight markers (lane M) that were identified and tracking dye (TD) are indicated by arrows. B, Molecular weight determination of UNG2 by matrix- assisted laser desorptionlionization mass spectrometer as described in Section 2.2.23. A mass spectrum was produced from 30 individual laser pulses. The relative intensity of charged ions is shown and mass determination was calculated using ion signals from an external calibrant in the same matrix. C, Ability of UNG2 to bind Ugi. Reaction mixtures (15 j.tl) containing 40 pmol of E. co/i Ung (lanes 2-3) or UNG2 (lanes 4-5) with (lanes 3 and 5) or without (lanes 2 and 4) Ugi (100 pmol) were incubated as described in Section 2.2.12.4. A control reaction containing Ugi (100 pmol) alone was similarly processed (lane 1). Samples were analyzed at 4°C using non-denaturing 10% polyacrylamide gel electrophoresis, proteins were visualized with Coomassie Brilliant Blue G-250 stain, and the gel was imaged as described in Section 2.2.12.4. Arrows indicate the location of Ung, UNG2, Ugi, UngUgi, UNG2Ugi and the tracking dye (TD). 1), Comparison of uracil-DNA glycosylase activity associated with UNG, UNG* and UNG2. Uracil-DNA glycosylase activity was measured under standard reaction conditions using 0.04-0.13 units of UNG*, UNG, or 1-4 units of UNG2 as described in Section 2.2.12.1. One unit of uracil-DNA glycosylase is defined as the amount of enzyme that releases 1 nmol of uracil per hour under standard conditions. The relative specific activity of UNG* and UNG2 are normalized to that of UNG, which was defmed as 100 %. Error bars represent the standard deviation of three experimental determinations. 195

MWx10 M II III IV V VI - 66.2-b- - - 45-*. - * - - 31-

21.5- * 14.4-i.- - TO-i.. B. 100

75 C 0) 50 >a) 25

0)

32 36 40 Mass (mlz x 10)

Lanes: 1 2 3 4 5

.._-UNG2 Ung--

higUgi-

-'Ugi TO-'- *

D.

50

50

UNG* UNG UNG2 Enzymes

Figure 30 I1'i.i

Figure 31. NaCI and MgC12 effects on recombinant UNG2 activity. A, NaC1 effect on UNG2 activity. Uracil-DNA glycosylase activity was measured under standard reaction conditions using 0.06 units of UNG (hued triangles) or UNG* (filled squares) and 1 units of UNG2 (filled circles) with 0-300 mM of NaC1 as described in Materials and Methods. The relative specific activities of enzyme in the presence of different NaC1 concentrations are normalized to that of enzyme without NaC1 addition, which was defmed as 100 %. Error bars represent the standard deviation of three experimental determinations. B, MgC12 effect on UNG2, UNG or UNG* activity. Uracil-DNA glycosylase activity was measured under standard reaction conditions using 0.06 units of UNG (fIlled triangles) or UNG* (fIlled squares) and 1 units of UNG2 (filled circles) with 0-15 mM of MgC12 as described above. The relative specific activities of enzyme in the presence of different MgC12 concentrations are normalized to that of enzyme without MgC12 addition, which was defined as 100 %. Error bars represent the standard deviation of three experimental determinations. One unit of uracil-DNA glycosylase is defined as the amount of enzyme that releases 1 nmol of uracil per hour under standard conditions. 197

A.

0 Li P

NaCI (mM)

B.

0

p

MgCl2 (mM)

Figure 31 198

Figure 32. NaC1 andMgCl2effects on UNG2 for binding to dU-containing double-stranded DNA. A, Changes in the 2AP fluorescence intensity of dU'2AP- 25-mer at 370 nm as a function of UNG2 concentration. di,U2AP-25-mer (50 riM) was mixed with 0-800 riM of UNG2 in buffer C (2 ml), and the fluorescence intensity of each addition was recorded at Xem=370 mn (X=3l0 rim) after mixing at 25°C for 5 mm. Data points represent the net fluorescence intensity, which was obtained by subtracting background protein fluorescence intensity from total sample fluorescence intensity at each concentration point. Data were a best fit to Equation 1 as described in

Section 2.2.18 (Kfor UNG2 = 270 ±35 nM). B, NaC1 effect on UNG2 (filled circles) or UNG (open circles) for binding to dijiU'2AP-25-mer. 50 nM of d1U.2AP-25-mer was mixed with 400 nM of UNG2 with 0-300 mM NaCl in buffer C (2 ml). The fluorescence intensity for each addition was recorded at 2em37° nm (kex=31O nm) after mixing at 25°C for 5 mm. Data points represent the net fluorescence intensity as described above. C,MgC12effect on UNG2 (filled circles) or UNG (open circles) for binding to dijrU.2AP-25-mer. 50 riM of dU.2AP-25-mer was mixed with 400 nM of UNG2 with 0-15 mMMgC12in buffer C (2 ml). The fluorescence intensity for each addition was recorded at Xem370 rim (A.ex310 rim) after mixing at 25°C for 5 mm. The displayed points represent the net fluorescence intensity which was obtained by subtracting out protein fluorescence intensity from sample fluorescence intensity at each concentration point. 199

A.

20

15 L5lb

200 400 600 800 UNG2 (nM) B. U20 15

U) 10

LLL. 5

0 100 200 300 NaCI (mM) U20 15

U) 10 0'-

0 4 8 12 16 MgCl2 (mM)

Figure 32 200

Figure 33. Pre-steady state kinetics of UNG2-induced fluorescence changes of 2AP containing oligonucleotide. A & B, stopped-flow time trace of UNG-induced fluorescence changes of 2AP-containing oligodeoxynucleotides was monitored using 2AP containing double-stranded oligodeoxynucleotide, dsT2AP-25-mer (A) or U2AP-25-mer (B) (200 nM), respectively, after mixing with UNG2 (ltM) in Buffer

C in a stopped-flow spectrometer as described inSection2.2.19. The 2AP fluorescence was excited at 310 nm and measured >350 tim. Each trace shown represents an average of 10 individual acquisitions. The solid lines represent the best- fit curves to Equation 3. C, Pre-steady state kinetics of UNG2-dsU'2AP-25-mer association was determined by monitoring the fluorescence enhancement of dsU2AP- 25-mer (200 nM) after mixing with 0.2-4 1iM of LJNG2. The rate constant obtained from each kinetic trace was plotted as a function of UNG2 concentration. The solid lines represent the best-fit curves to Equation 4 as described inSection 2.2.20. D, NaC1 effect on UNG2-induced 2AP fluorescence enhancement. The NaC1 effect on pre-steady state kinetics of UNG2-dsU2AP-25-mer association was determined by monitoring the 2AP fluorescence change of dsU2AP-25-mer (200 nM) after mixing with UNG2 (1 riM) containing 0-150 mM of NaC1 in buffer C. 201

A.

Time (Seconds) B. 0.005 5) 0.000 0.005

0.70

5)__ 065 o

U. 050

Time (Seconds) C.

- 450

300 0 150

2 3 4 UNG2(M) D.

450 300) .0 0 150

Sf) IflO NaCI(mM)

Figure 33 202

Figure 34. Effect of MgC12 on the observed rate of UING2- and UNG-induced 2AP fluorescence enhancement. The effect of MgCl2 on the pre-steady state kinetics of dsU2AP-25-mer- binding by TJNG2-(filled circles)or UNG-(open squares)was determined by monitoring the 2AP fluorescence change of dsU2AP-25-mer (200 nM) after mixing with UNG2 (1 jtM) containing 0-16 mM of MgC12 in buffer C. Fluorescence measurements and determination of apparent rates were carried out as described in Figure 33 legends. 203

450

300

- 150

MgCl2 (mM)

Figure 34 204

Figure 35. Stopped-flow time trace of UNG2 intrinsic protein fluorescence upon binding duplex uracil-DNA. The intrinsic protein fluorescence of UNG2 (20 nM) was monitored in a stopped-flow spectrometer after mixing with 1 iM dsTA-25-mer (A) or dsUA-25-mer (B). Each kinetic trace represents an average often individual acquisitions. The solid lines represent the best-fit curves to Equation 2. B, top panel: residual analysis of each trace to the fit curve. C, the intrinsic protein fluorescence change of UNG2 (200 nM) as monitored in a stopped-flow spectrometer after mixing with 0.2-2 tM dsUA-25-mer, respectively. The observed rate constants of individual kinetic trace were obtained by fitting each trace to Equation 3 as described in Section 2.2.20. The observed rate constants of individual kinetic trace were obtained by fitting each trace to Equation 2 as described in Section 2.2.20. The filled squares and filled triangles represent the rates of rapid (kQa) and slow (kQb) fluorescence quenching, respectively. The open circles represent the rate of slow fluorescence recovery (kR). 205

A. U0) 1.2

I!::

1 0.2 0.3 0.4 Time (Seconds) B.

U03 C 03.- $Po 0I-

Time (Seconds) C.

I.. cn 200F

0) .0 0 .x 100

0.5 1 1.5 LJA-25-mer (gM)

Figure 35 206

Figure 36. The effect of NaC1 on the transient change in UNG2 intrinsic fluorescence associated with binding dsUA-25-mer. A, UNG2 (100 nM) was mixed with 1 j.tM dsUA-25-mer containing 0, 25, 50, 75, 100, or 150 mM of NaC1 in buffer C as indicated and the fluorescence change was monitored in a stopped flow spectrometer. Each stopped-flow trace shown represents an average of 10 individual acquisitions. The solid lines represent the best-fit curves to Equation 2. B, the observed rate constants for fluorescence quenching(kQaandkQb,fihled squaresand filled triangles,respectively) and fluorescence recovery(kR, opencircles) were obtained for the reactions described in (A) and graphed as a function of NaC1 concentration. A.

+25 mM NaCI

+50mM NaCI >0 0.8

0) 0.8 C.)

0.8 +75 mM NaCI 0) C) I- 0

+100 mM NaCI 0.8

+150 mM NaCI

0.6

0:2 0:3 Time (Seconds)

(0 .0U, 0

NaCI (mM)

Figure 36 208

Figure 37. Effect of MgCJ2 on the transient change in UNG2 intrinsic fluorescence associated with binding dsUA-25-mer. UNG2 (100 nM) was mixed with 1 j.M dsU.A-25-mer containing 0, 1, 2,4, 8, 16, or 32 mM ofMgC12and the change in intrinsic fluorescence was monitored in a stopped flow spectrometer. Each stopped-flow trace shown represents an average of 10 individual acquisitions. The solid lines represent the best-fit curves to Equation 2. The observed rate constants for each stopped-flow trace were shown in Figure 38. 209

+0 mMMgCl2

I

IVTTT +1 mMMgCl2 I:TT +2 mMMgCl2 (1)1 0

C, +4 mMMgCl2

C) C.) ("I: 0 +8 mMM9CJ2 .8

U-

I.4 I I +16 mMMgCl2

I.8

1.4 I I +32 mMMgCl2

I.8

I.4 0.1 0.2 0.3 0.4 Time (Seconds)

Figure 37 210

Figure 38. Observed rate constants for the dsUA-25-mer-induced intrinsic protein fluorescence change of IJNG2 or UNG in the presence of MgCl2. The observed rate constants associated with UNG2 (A) and UNG (B) intrinsic fluorescence change on binding dsUA-25-mer are shown. The filled squares and filled triangles represent the rates of rapid (kQa) and slow (kQb) fluorescence quenching, respectively. The open circles represent the rate of initial fluorescence recovery (k). 211

A.

0

0 8 16 24 32 MgCl2 (mM)

300

200

2 0 100

MgCl2 (mM)

Figure 38 212

Figure 39. Direct comparison of stopped-flow time traces of 2AP fluorescence enhancement and intrinsic UNG2 fluorescence quenching using dsU'A-25-mer and ds-'UA-25-mer. Stopped-flow time traces of 2AP fluorescence enhancement were obtained by mixing with either 200 nM of dsU2AP-25-mer (A, top panel) or ds- dU'2AP-25-mer (B, top panel) with 1 .iM of UNG2 in buffer C. Time traces of intrinsic UNG2 fluorescence quenching were obtained by mixing with either 1 pM of dsU'A-25-mer (A, bottom panel) or ds-dijiU.A-25-mer (B, bottom panel) with 200 tiM of IJNG2 in buffer C. Each trace shown represents an average of 10 individual acquisitions. The solid lines in top panels represent the best-fit curves to Equation 3 as described in Section 2.2.20. The solid lines in bottom panels represent the best-fit curves to Equation 2 as described in Section 2.2.20. Direct comparison of two traces was obtained at equal molar concentrations of enzyme to minimize concentration- dependent effects. 213

A.

2.

2.

>

, 2.

C.)

a) C., U)

22

LI.. zi

2.0

u.ub 0.10 0.15 Time (Seconds)

2

2.

>

a)

1.6

Time (Seconds)

Figure 39 214

6. Comparison of the Nucleotide Flipping Mechanism ofEscherichia coli and Human Uracil-DNA Glycosylase

This chapter presents studies ofE.coil Ung and human UNG nucleotide flipping and protein conformational changes designed to address the mechanistic

question of temporal succession: does pull occur before push, orvice versa. A transient kinetics approach was utilized to monitor 2AP fluorescence change resulting from Ung and UNG binding to fluorescent dsU2AP-25-mer or ds-ijiU2AP-25mer DNA, and Ung and UNG protein fluorescence change associated with binding non- fluorescent dsU'A-25-mer and ds-UA-25-mer DNA. The stopped-flow time traces of 2AP and intrinsic protein fluorescence change were compared side by side to determine whether nucleotide flipping or protein conformational change occurred first.

The results forE.coil Ung show that the enzyme-induced 2AP fluorescence enhancement (nucleotide flipping) occurred prior to the quenching of intrinsic protein fluorescence (conformational change) observed when enzyme reacted with a duplex uracil-containing DNA. However, whenE.coil Ung was reacted with PU-containing DNA, the two fluorescent signals happened almost simultaneously. Similar results were observed for human UNG in reactions that contained uracil-DNA. However, when UNG was reacted with iiU-containing DNA, the intrinsic protein fluorescence remained unchanged. Another distinction between theE.coil Ung and human UNG reaction mechanisms was observed in reactions containing ssU- or U-25-mer DNA. Quenching of Ung intrinsic protein fluorescence upon binding these single-stranded DNAs did not recover to initial levels during the 8 second stopped-flow time scale. In contrast, this delayed fluorescence recovery was not observed in UNG-single-stranded DNA interactions. These results presented in this Chapter demonstrate that both Ung and UNG employ a "Pull-Push" mechanism for uracil flipping from duplex DNA. However, Ung switches to a "Push-Pull" mechanism when binding duplex jiU- containing DNA. 215

6.1. Results

6.1.1. Fluorescence Properties of2AP-containing Oligonucleotides

A double-stranded U2AP-25-mer, containing a site-specific U residue opposite 2AP at position 11 from the 5'-end, was constructed in order to study the binding and uracil-flipping activity in the pre-steady state. The fluorescence properties of the dsU2AP-25-mer were monitored as described in Section 2.2.17. As shown in Figure 40B, annealing 400 nM of single-stranded (ss) 2AP-25-mer with increasing concentrations (0-800 nM) of the complementary ssU-25-mer resulted in an incremental reduction of fluorescence (Figure 40B, af). Addition of excess ssU-25- mer did not produce further fluorescence decreases (Figure 40B, g). The fluorescence excitation and emission spectra of 400 nM dsU2AP-25-mer exhibited the characteristic 2AP fluorescence excitation and emission maxima of -310 nm and 370 nm, respectively (Figure 40A). These results were consistent with those reported by Wong et al. (121), and indicated the duplex U'2AP-25-mer would be a useful probe to detect uracil flipping.

6.1.2. Pre-steady State Kinetic Analysis of Uracil Fl4pingby (JNG

To examine the pre-steady state kinetics of UNG DNA binding and uracil flipping, stopped-flow experiments were performed using 100 nM dsU2AP-25-mer and 1 j.tM UNG as described in Section 2.2.19. As shown in Figure 41B, the 2AP fluorescence trace showed a rapid and transient enhancement within the first 5 ms. This 2AP fluorescence enhancement was UNG-dependent, since the fluorescence change was not observed in the reaction without UNG addition (Figure 41A). The stopped-flow trace was fitted to Equation 3, and the observed rate of 2AP fluorescence enhancement (kfl1p) was determined to be .525 s1. The dependence of the observed rate on UNG concentration was obtained by titration of 100 nM of dsU2AP-25-mer 216

with 200-2400 nM of UNG; the rate constants ranged from 92 to 620 s_i (Figure 41C). The non-linear dependence of the observed rate constants on UNG concentrations ruled out a direct one-step uracil flipping mechanism concomitant with DNA binding.

These results were similar to that for UNG2 (Figure 33 C), but the maximumkfli for UNG was faster than that for IJNG2 (620 s_i vs. 450 s1).

6.1.3. Direct Comparisonof2AP and Intrinsic Protein Fluorescence Change When Ung or UNG Binds a Double-stranded Uracil- containing DNA

Previously, Wong et al. (121) demonstrated that the 2AP fluorescence enhancement induced by E. coli Ung binding to duplex uracil-DNA occurred earlier than the change in intrinsic protein fluorescence. As demonstrated in the Chapter Five of this Dissertation, the 2AP fluorescence enhancement induced by TJNG2 binding to duplex uracil-DNA also occurred earlier than the change in UNG2 intrinsic fluorescence. There results indicated that uracil nucleotide flipping occurred earlier than Ung or UNG2 conformational change. To examine whether uracil nucleotide flipping took place before conformational change in reactions containing UNG, stopped-flow experiments were performed as described in Chapter Five. Reactions contained either a 10-fold excess of dsUA-25-mer DNA to enzyme (UNG or E. coil Ung), or a 10-fold excess of enzyme to DNA, in order to minimize concentration- dependent effects. As shown in Figure 42A, enhancement of 2AP fluorescence induced byE. coli Ung (top panel) occurred prior to the change in Ung intrinsic fluorescence (bottom panel). The maximum of 2AP fluorescence enhancement occurred at -'9 ms, while maximal quenching of E. coil Ung intrinsic fluorescence occurred at25 ms. Data analysis confirmed that the rate of 2AP fluorescence enhancement (kfljp-380 1)was more rapid than the fast rate of enzyme intrinsic fluorescence quenching(kQa =-'200 s'). These results were consistent with those reported by Wong et al. (121). Similar results were observed for UNG. As shown in Figure 42B, the maximum of 2AP fluorescence enhancement occurred at8 ms (top panel), while maximal quenching of E. coli Ung intrinsic fluorescence occurred at 217

13 ms (bottom panel). Data analysis revealed that the rate of 2AP fluorescence

enhancement (kfl = -'430s'1)was more rapid than the fast rate of enzyme intrinsic fluorescence quenching(kQa --230 s4). Thus, like E. coil Ung and UNG2, uracil- flipping preceded the UNG conformational change, indicating that the absence of the UNG2 N-terminal domain did not affect the mechanism of T.JNG.

6.1.4. Comparison of UNG andE. coil Ung 2AP Fluorescence Enhancement and Intrinsic Protein Fluorescence Change Using dyiU2AP-25-mer DNA

In Chapter Five, equilibrium DNA binding/uracil-flipping measurements showed that while TJNG2 bound non-hydrolyzable VU-containing DNA (Figure 32A), quenching of UNG2 intrinsic protein fluorescence was not observed (Figure 39B, bottom panel).To determine whether VU-containing DNA would have a similar effect on the intrinsic fluorescence of E. coli Ung and UNG, stopped-flow experiments were conducted as described in Section 6.1.3, except that fluorescent dsUAE-25-mer and non-fluorescent dsU'A-25-mer were replaced by ds-dU-25-mer and ds-dijiU.A- 25-mer, respectively. As shown in the Figure 43A (top panels), the amplitude of 2AP fluorescence enhancement by E. coil Ung was reduced by 50% as compared to that of Figure 42A (top panels). In addition, the maximum 2AP fluorescence enhancement occurred at10 ms (Figure 43A, top panels), which was slower than that (-- 9 ms) of Figure 42A (top panels). Similar results were observed for UNG (Figures 42B and 43B, bottom panels). As shown in the Figure 43B (top panels), the maximum 2AP fluorescence enhancement occurred at -9 ms, which was slower than that ('- 8 ms) of Figure 42B (top panels). Thus, the rate of base flipping was slower when E. coli Ung or UNG bound duplex iU-containing DNA. In contrast, as shown in Figure 43A (bottom panel) both the rate and amplitude of Ung protein fluorescence quenching were similar to those obtained with uracil-DNA (Figure 42A, bottom panels). However, the fluorescence recovery was not observed when the enzyme was reacted with duplex ijU-containing DNA. Furthermore, the intensity of UNG intrinsic fluorescence remained unchanged when the enzyme bound the duplex ijiU-containing 218

DNA (Figure 43B, bottom pane!). These results demonstrate that the non-hydrolyzable duplex PU-containing DNA trapped Ung in the closed conformation for at least 10 seconds, and that UNG could not undergo conformational change when bound to duplex NJU-DNA. Direct comparison of the stopped-flow time traces of 2AP fluorescence enhancement (Figure 43A, top panel) and Ung intrinsic protein fluorescence quenching (Figure 43A, bottom pane!) revealed that 2AP and protein fluorescence change occurred almost simultaneously. Data analysis showed that the for E. coli

1) Ung (238 was close to the kQaof E. coli Ung (- 230 s1). These results suggested that uracil-flipping occurred concurrently with enzyme conformational change when Ung bound duplex JU-containing DNA.

6.1.5. Direct Comparison of E. coli Ung and UNG Intrinsic Fluorescence Change Induced by Binding to Single-stranded Uracil-containing DNA

As observed in Chapter 4, the rate of UNG intrinsic fluorescence change associated with binding single-stranded uracil-DNA was faster than with duplex uracil-DNA (Figure 28). To compare further the change in E. coli Ung and UNG intrinsic fluorescence induced by single-stranded uracil-DNA binding, stopped-flow experiments were performed using 100 nM of E. coli Ung or UNG and 1 jiM of ssU- 25-mer. As shown in Figure 44B, the stopped-flow time trace of UNG intrinsic fluorescence change showed a rapid fluorescence quenching in the first 10 milliseconds, followed by slow fluorescence recovery that lasted up to -200 ms. This kinetic behavior resembled that obtained with dsU'A-25-mer (Figure 23). Data analysis confirmed that the rate (-'380 1)of fluorescence quenching associated with binding ssDNA (Figure 44B) was faster than that (-195 s1) with duplex DNA (Figure 42B, bottom panel). However, as shown in Figure 44A, the stopped-flow time trace of E. coli Ung intrinsic fluorescence change associated with binding ssDNA displayed a rapid fluorescence quenching in the first 8 milliseconds, and remained unchanged for up to 8 seconds. Then, the fluorescence recovered to its initial intensity within another 219

8 milliseconds. Like UNG, the E. coil Ung intrinsic fluorescence change associated with binding ssDNA was also uracil-DNA specific, since no fluorescence change was observed when Ung (100 nM) reacted with 1 p.M of non-specific ssT-25-mer. Data analysis disclosed that the fast rate of E. coil Ung intrinsic fluorescence quenching, kQa, (Figure 44A, bottom panel) was faster for single-stranded DNA (-250 s1) than for duplex DNA (-'200s1)(Figure 42A, bottom panel). These results demonstrated that, forE. coli Ung, the conformational change associated with binding single-stranded uracil-DNA was faster than that observed with binding duplex uracil-DNA.

6.1.6. Direct ComparisonofE. coli Ung and UNG Intrinsic Fluorescence Change Induced by Binding to Single-stranded VU-DNA

Finally, to investigate whether a difference existed between E. coil Ung and UNG intrinsic fluorescence change associated with binding ss-U-25-mer, stopped- flow experiments were performed using 100 nM of E. coil Ung or UNG reacted with 1 p.M of ss-iiJU-25-mer. As shown in Figure 45A, the observed stopped-flow time trace of Ung intrinsic fluorescence change showed a steady decrease in fluorescence with time. This result was quite different from that with ds-UA-25-mer (Figure 45A, bottom panel). Unlike E. coil Ung, a change in UNG intrinsic fluorescence was not observed with ss-U-25-mer (Figure 39B), or with ds-UA-25-mer (Figure 45B, bottom panel). 220

Figure 40. Fluorescence properties of 2AP-containing oligonucleotides. A, fluorescence excitation (dashed line) and emission (solid line) spectra of double- stranded U2AP-25-mer in buffer A. 400 nM single-stranded 2AP-25-mer was annealed with 440 nM single-stranded U-25-mer in buffer A. The annealing reaction is described in Section 2.2.17. The real-time fluorescence emission spectra was recorded atXenl 370nm (X=310mu) at 25°C. B, quenching of 2AP fluorescence upon mixing with single-stranded U-25-mer. Single-stranded U-25-mer was added to 400 nM 2AP-25-mer in buffer A at 0 (dash line), 50, 100,200,300,400,500, and 600 nM (a-g, respectively). The real-time fluorescence emission spectra for each addition was recorded at2eni370muQ'ex310nm) after mixing at 25°C for 10mm.Arrows indicate the wavelength at peak fluorescence. 221

A.

25 37Onm

20

15

250 300 350 400450 500 Wavelength (nm) B. 80 370 nm a-

'40 jL20

350 400 450 500 Wavelength (nm)

Figure 40 222

Figure 41. Pre-steady state kinetic analysis of uracil flipping induced byIJNG binding. A & B, stopped-flow time traces of UNG-induced 2AP fluorescence change was recorded using 100 nM of T'2AP-25-mer (A) or U2AP-25-mer (B) mixing with 1 .iM of UNG in Buffer A in Stopped-flow reaction analyzer as described in Section

2.2.19. The 2AP fluorescence was excited at)ex= 331 nm and measured Xem> 350 nm. Each kinetic trace shown represents an average of 10 individual acquisitions. The solid lines represent the best-fit curves to Equation 3 as described inSection 2.2.20. C, Stopped-flow kinetics of association of UNG and U2AP-25-mer was determined by monitoring the fluorescence enhancement of U2AP-25-mer (100 nM) after mixing with UNG at 200, 250, 300, 400, 500, 600, 800, 1200, 1600,2000, and 2400 nM, respectively. The rate constant obtained from each acquisition was plotted as function of UNG concentration. The solid lines represent the best-fit curves to Equation 4 as described inSection 2.2.20. 223

A. 0 Z4 0 Z2______gzo

0.005 010 0.015 Time (Seconds) B. 0 0

0 U o

I-00

o.

0.005 0.010 0.015 Time (Seconds)

800 ;600 C .c 400 0 200

UNG (mM)

Figure 41 224

Figure 42. Comparison of UNG 2AP fluorescence enhancement and intrinsic protein fluorescence change with E. coil Ung using duplex uradil-DNA. Comparable time courses of 2AP and intrinsic tryptophan fluorescence stopped-flow traces were obtained by mixing with either 1 .iM of E. coli Ung (A) orUNG(B) and 100 nM of dsU2AP-25-mer or 100 nM of E. coli Ung (A) orUNG(B) and 1 M of dsUA-25-mer in buffer A as described in Section 2.2.19. Each trace shown represents an average of 10 individual acquisitions. The solid lines represent the best-fit curves to Equation 2 (bottom panels) and 3 (top panels) as described in Section 2.2.20. 225

A.

2.0

I TI

0.005 0.015 0.025 Time (Seconds) B.

0a) C 0

0

U.

0.005 0.015 0.025 Time (Seconds)

Figure 42 226

Figure 43. Comparison of 2AP fluorescence enhancement and intrinsic protein fluorescence quenching induced by Ung and UNG binding to duplex PU-DNA. Comparable stopped-flow time traces of 2AP fluorescence enhancement and protein intrinsic fluorescence change were obtained by mixing either 1 M of Ung (A) or UNG (B) with 100 nM of ,U2AP-25-mer or 100 tiM of Ung (A) or UNG (B) with 1 tM of U2AP-25-mer in buffer A as described in Section 2.2.19. Each kinetic trace shown represents an average of 10 individual acquisitions. The solid lines represent the best-fit curves to Equation 2 as described in Section 2.2.20 for protein fluorescence trace or Equation 3 as described in Section 2.2.20 for 2AP fluorescence trace. Direct comparison of two traces was obtained at equal molar concentrations of enzyme to minimize concentration-dependent effects. 227

A.

0.9 0C) 0C) U) 0

LL

0.005 0.015 0.025 Time (Seconds) B.

1.3 I

1.1

0C) U,

2.0

I.

1.8

0.005 0.015 0.025 Time (Seconds)

Figure 43 228

Figure 44. Comparison of the change in intrinsic protein fluorescence of E. coli Ung and UNG associated with binding single-stranded uracil-DNA. Comparable stopped-flowed time traces of protein intrinsic fluorescence change were obtained by mixing either 100 nM of E. coli Ung (A) or UNG (B) with 1 jtM of ssU-25-mer in buffer A as described inSection 2.2.19.Each kinetic trace shown represents an average of 10 individual acquisitions. The solid lines represent the best-fit curves to

Equation 2 as described inSection 2.2.20.Direct comparison of the two traces was obtained at equal molar concentrations of enzyme to minimize concentration- dependent effects. 229

A.

1.5 0 In 0

LI.. 0.05 0.10 0.15 0.20 Time (Seconds) I 0

0

LI- 0.05 0.10 0.15 0.20 Time (Seconds)

Figure 44 230

Figure 45. Comparison of the change in intrinsic protein fluorescence of Ung and UNG associated with binding single-stranded VU-DNA. Comparable stopped- flowed time traces of protein intrinsic fluorescence change were obtained by mixing either 100 nM of Ung (A) or UNG (B) with 1 M of ss-U-25-mer in buffer A as described inSection 2.2.19.Each kinetic trace shown represents an average of 10 individual acquisitions. The solid lines represent the best-fit curves to Equation 2 as described inSection 2.2.20.Direct comparison of two traces was obtained at equal molar concentrations of enzyme. 231

A.

1.7

0 0 1.6

1.5 0.05 0.10 0.150.20 Time (Seconds) B. 41.7

0 0 1.6 0 U, 0 1.5 U. 0.05 0.100.15 0.20 Time (Seconds)

Figure 45 232

6.2. Discussion

Direct comparison of the stopped-flow time traces of 2AP fluorescence andE. co/i Ung intrinsic fluorescence change associated with binding duplex uracil-DNA showed that the 2AP fluorescence enhancement occurred before the quenching of intrinsic protein fluorescence. These results suggested that uracil nucleotide flipping occurredbefore E.co/i Ung conformational change. This observation was consistent with the previous report by Wonget al(121). Similar results were obtained for UNG.

These results suggested that bothE.co/i Ung and UNG implemented a "pull-push" mechanism to flip and bind the uracil nucleotide. Hence, it is possible that uracil nucleotide flipping precedes enzyme conformational change is a common mechanistic phenomenon for family-i uracil-DNA glycosylases. However, when E. coli Ung was reacted with ds-U'2AP-25-mer DNA, 2AP fluorescence enhancement and enzyme intrinsic fluorescence quenching occurred concurrently (Figure 40A). These results indicated that i1jU nucleotide flipping andE.co/i Ung conformational change happened at the same time. Thus,E.coli Ung utilized a simultaneous pushing and pulling mechanism for flipping and binding a non-hydrolyzable analog of deoxyuridine-DNA. The 2AP fluorescence enhancement was uncoupled with UNG intrinsic fluorescence change when UNG reacted with ds-ijjU'2AP-25-mer DNA (Figure 43). This result suggested that ijjU nucleotide flipping still occurred, but UNG did not undergo a conformational change. In the UNG*NjUDNA co-crystal structure, binding to the UNG* uracil-specificity pocket rotated the uracil ring by900 on its Ni-C4 axis from its normal anti-conformation to a position halfway between anti and syn configuration (Figure 3) (117). Since the U base was still connected to the deoxyribose in the enzyme-DNA complex, the iU deoxyribose was pulled0.4A deeper into the enzyme active site pocket than in the enzyme-product complex (116,117). However, the U base was sterically prevented from inserting as deeply into the active site center as the cleaved uracil in the enzyme-product complex. 233

Therefore, the stacking interaction between the U base and Phe 158 aromatic side chain was less than ideal. Analogously, the hydrogen bonding between U 02 and 04, and the side chains of T.JNG* H268 and N204 would be weaker than those observed for uracil, as deduced from the structure of the enzyme-product complex (116). The altered interactions between the U base and the enzyme active site residues may be insufficient to trigger enzyme isomerization to the closed conformation. If the enviromnent of W245 on 3 (Figure 24B) is not affected by NJU binding, and W245 is indeed the reporting fluorophore for stably bound uracil, then the UNG intrinsic fluorescence will remain unchanged. A similar argument can be made for UNG binding to ss-U-25-mer. However, the active site geometry ofE. coil Ung and the environment of the putative fluorescent reporter, W163, may be subtly different, such that quenching of E. coil Ung intrinsic fluorescence was observed in reactions containing ds-U2AP-25-mer DNA.

The stopped-flow time trace ofE. coilUng intrinsic fluorescence change associated with binding ss-U-25-mer showed that the enzyme fluorescence did not recover for up to 8 seconds. This result indicated that theE. coilUng remained in a closed conformation, and was likely bound to the ssDNA reaction products, AP-site- 25-mer and the cleaved uracil base. In addition, this delayed product release was only observed in the reactions with ss-U-25-mer. It is likely thatE. coilUng remained bound to the single stranded AP-site-containing DNA after the glycosylic bond was cleaved. In fact, family-i uracil-DNA glycosylases have a high affinity for AP-sites (57). For example, mitochondrial uracil-DNA glycosylase from rat liver bound uracil-

DNA with an apparentKmof 1.1 jiM, and AP-site DNA with aK1of 1.2 p.M (87).

The stopped-flow time trace ofE. co/iUng intrinsic fluorescence change associated with binding ss-ijiU-25-mer showed that rate of enzyme fluorescence quenching was much slower than the rate of quenching observed for binding to ds- U.A-25-mer. This result suggested that the uracil-specificity pocket of E. coil Ung slowly adjusted its conformation to accommodate the U base. The rapid rate ofE. coil Ung intrinsic fluorescence quenching associated with binding ds-iU.A-25-mer 234 indicated that the complementary strand was necessary for efficient conformational change. It is possible that in single-stranded DNA, where the DNA sugar-phosphate backbone is more flexible than in double-stranded DNA, the non-standard geometry of the deoxypsuedouridine nucleotide is difficult to fit into the open conformation of the active site. Moreover, in single-stranded DNA, the leucine 191-side chain is ineffective in "pushing" the deoxypsuedouridine nucleotide into the active site pocket. However, in double-stranded DNA, the orientation of the non-standard U nucleotide is constrained by base pairing and base stacking interactions with other deoxynucleotides in the double helix. These interactions compel the U nucleotide to adopt a more standard conformation. Consequently, the enzyme participates in positive intereactions with the NJU nucleotide and the complementary strand, which enable leucine 191 to flip ("push") the non-standard baseintothe open active site and prevent ("plug") its return to the DNA helix. The experiments with U suggest that glycosylic bond cleavage, leucine loop retraction, and product release must occur before the enzyme can return to the open conformation. BIBLIOGRAPHY

1. Das, S. K., Kunkel, T. A., and Loeb, L. A. (1985) Effects of altered nucleotide concentrations on the fidelity of DNA replication. Basic Ljfe Sci 31, 117-126.

2. Brynolf, K., Eliasson, R., and Reichard, P. (1978) Formation of Okazaki fragments in polyoma DNA synthesis caused by misincorporation of uracil. Cell 13, 573-580.

3. Tye, B. K., Chien, J., Lehman, I. R., Duncan, B. K., and Warner, H. R. (1978) Uracil incorporation: a source of pulse-labeled DNA fragments in the replication of the Escherichia coli chromosome. Proc NatlAcadSci USA 75, 233 -23 7.

4. Burgers, P. M., and Klein, M. B. (1986) Selection by genetic transformation of a Saccharomyces cerevisiae mutant defective for the nuclear uracil-DNA- glycosylase. JBacteriol 166, 905-9 13.

5. Goulian, M., Bleile, B., and Tseng, B. Y. (1980) Methotrexate-induced misincorporation of uracil into DNA. Proc NatlAcadSci USA 77, 1956-1960.

6. Goulian, M., Bleile, B., and Tseng, B. Y. (1980) The effect of methotrexate on levels of dUTP in animal cells. JBiol Chem 255, 10630-10637.

7. Olivera, B. M. (1978) DNA intermediates at the Escherichia coli replication fork: effect of dUTP. Proc NatlAcad Sci USA 75, 238-242.

Dube, D. K., Kunkel, T. A., Seal, G., and Loeb, L. A. (1979) Distinctive properties of mammalian DNA polymerases. Biochim Biophys Acta 561, 369- 382.

9. Tye, B. K., Nyman, P. 0., Lehman, I. R., Hochhauser, S., and Weiss, B. (1977) Transient accumulation of Okazaki fragments as a result of uracil incorporation into nascent DNA. Proc NatlAcadSci USA 74, 154-157.

10. Focher, F., Mazzarello, P., Verri, A., Hubscher, U., and Spadari, S. (1990) Activity profiles of enzymes that control the uracil incorporation into DNA during neuronal development. Mutat Res 237, 65-73.

11. Yoshida, S., and Masaki, S. (1979) Utilization in vitro of deoxyuridine triphosphate in DNA synthesis by DNA polymerases alpha and beta from calf thymus. Biochim Biophys Acta 561, 396-402. 236

12. Mosbaugh, D. W. (1988) Purification and characterization of porcine liver DNA polymerase gamma: utilization of dUTP and dTTP during in vitro DNA synthesis. Nucleic Acids Res 16, 5645-5659.

13. el-Hajj, H. H., Zhang, H., and Weiss, B. (1988) Lethality of a dut (deoxyuridine triphosphatase) mutation in Escherichia coli. JBacteriol 170, 1069-1075.

14. O'Donovan, G. A., Edlin, G., Fuchs, J. A., Neuhard, J., and Thomassen, B. (1971) Deoxycytidine triphosphate deaminase: characterization of an Escherichia coli mutant deficient in the enzyme. JBacteriol 105, 666-672.

15. Beck, C. F., Eisenhardt, A. R., and Neuhard, J. (1975) Deoxycytidine triphosphate deaminase of Salmonella typhimurium. Purification and characterization. JBiol Chem 250, 609-6 16.

16. Reichard, P. (1988) Interactions between deoxyribonucleotide and DNA synthesis. Annu Rev Biochem 57, 349-3 74.

17. Mathews, C. K. (1993) Enzyme organization in DNA precursor biosynthesis. Frog Nucleic Acid Res Mol Biol 44, 167-203.

18. Lundberg, L. G., Thoresson, H. 0., Karistrom, 0. H., and Nyman, P. 0. (1983) Nucleotide sequence of the structural gene for dUTPase of Escherichia coli K- 12.EmboJ2, 967-971.

19. Cedergren-Zeppezauer, E. S., Larsson, G., Nyman, P. 0., Dauter, Z., and Wilson, K. S. (1992) Crystal structure of a dUTPase. Nature 355,740-743.

20. Shlomai, J., and Kornberg, A. (1978) Deoxyuridine triphosphatase of Escherichia coli. Purification, properties, and use as a reagent to reduce uracil incorporation into DNA. JBiol Chem 253, 3305-3312.

21. Hochhauser, S. J., and Weiss, B. (1978) Escherichia coli mutants deficient in deoxyuridine triphosphatase. JBacteriol 134, 157-166.

22. Warner, H. R., Duncan, B. K., Garrett, C., and Neuhard, J. (1981) Synthesis and metabolism of uracil-containing deoxyribonucleic acid in Escherichia coli. JBacteriol 145, 687-695.

23. Blount, B. C., and Ames, B. N. (1994) Analysis of uracil in DNA by gas chromatography-mass spectrometry. Anal Biochem 219, 195-200. 237

24. Blount, B. C., Mack, M. M., Wehr, C. M., MacGregor, J. T., Hiatt, R. A., Wang, G., Wickramasinghe, S. N., Everson, R. B., and Ames, B. N. (1997) Folate deficiency causes uracil misincorporation into human DNA and chromosome breakage: implications for cancer and neuronal damage. Proc NatlAcadSci USA 94, 3290-3295.

25. Duthie, S. J., and McMillan, P. (1997) Uracil misincorporation in human DNA detected using single cell gel electrophoresis. Carcinogenesis 18, 1709-1714.

26. Duncan, B. K., and Miller, J. H. (1980) Mutagenic deamination of cytosine residues in DNA. Nature 287, 560-561.

27. Duncan, B. K., and Weiss, B. (1982) Specific mutator effects of ung (uracil- DNA glycosylase) mutations in Escherichia coli. JBacteriol 151, 750-755.

28. Shapiro, R., and Klein, R. S. (1966) The deamination of cytidine and cytosine by acidic buffer solutions. Mutagenic implications. Biochemistry 5, 2358-2362.

29. Shapiro, R., Braverman, B., Louis, J. B., and Servis, R. E. (1973) Nucleic acid reactivity and conformation. II. Reaction of cytosine and uracil with sodium bisulfite. JBiol Chem 248,4060-4064.

30. Hayatsu, H. (1976) Bisulfite modification of nucleic acids and their constituents. Prog Nucleic AcidRes MolBiol 16, 75-124.

31. Lindahl, T., and Nyberg, B. (1974) Heat-induced deamination of cytosine residues in deoxyribonucleic acid. Biochemistry 13, 3405-34 10.

32. Lindahl, T. (1979) DNA glycosylases, endonucleases for apurinic/apyrimidinic sites, and base excision-repair. Prog Nucleic Acid Res MolBiol 22, 135-192.

33. Frederico, L. A., Kunkel, T. A., and Shaw, B. R. (1990) A sensitive genetic assay for the detection of cytosine deamination: determination of rate constants and the activation energy. Biochemistry 29, 2532-2537.

34. Verri, A., Mazzarello, P., Spadari, S., and Focher, F. (1992) Uracil-DNA glycosylases preferentially excise mispaired uracil. Biochem J 287 (Pt 3), 1007-1010.

35. Mitchell, D. L., Jen, J., and Cleaver, J. E. (1991) Relative induction of cyclobutane dimers and cytosine photohydrates in DNA irradiated in vitro and in vivo with ultraviolet-C and ultraviolet-B light. Photochem Photobiol 54, 741-746. 238

36. Shaw, A. A., Falick, A. M., and Shetlar, M. D. (1992) Photoreactions of thymine and thymidine with N-acetyltyrosine. Biochemistry 31, 10976-10983.

37. Grossman, L., and Rodgers, E. (1968) Evidence for the presence of cytosine photohydrates in UV irradiated nucleic acids. Biochem. Biophys. Res. Commun. 33, 975-983.

38. Patrick, M. H., and Rahn, R. 0. (1976) in Photochemistry and Photobiology of Nucleic Acids (Wang, S. Y., ed) Vol. 2, pp. 3 5-95, Academic Press, New York, N.Y.

39. Lecomte, P., Boiteux, S., and Doubleday, 0. (1981) Mechansim of ultraviolet- induced mutagenesis: the coding properties of ultraviolet-irradiated poly(dC) replicated byE. coli DNA polymerase I. Nucleic Acids Res. 9, 3491-3501.

40. Fisher, G. J., and Johns, H. E. (1976) Photochemistiy and photobiology of nucleic acids, 1, Academic Press, New York

41. Boorstein, R. J., Hubert, T. P., Cunningham, R. P., and Teebor, G. W. (1990) Formation and stability of repairable pyrimidine photohydrates in DNA. Biochemistry 29, 10455-10460.

42. Boorstein, R. J., Hubert, T. P., Cadet, J., Cunningham, R. P., and Teebor, G. W. (1989) UV-induced pyrimidine hydrates in DNA are repaired by bacterial and mammalian DNA glycosylase activities. Biochemistry 28, 6164-6170.

43. Setlow, R. B., Carrier, W. L., and Bollum, F. J. (1965) Pyrimidine dimers in UV-irradiated poly dI:dC. Proc. Nail. Acad. Sci. U.S.A. 53, 1111-1118.

44. Peng, W., and Shaw, B. R. (1996) Accelerated deamination of cytosine residues in UV-induced cyclobutane pyrimidine dimers leads to CC to TT transitions. Biochemistry 35, 10172-10181.

45. Barak, Y., Cohen-Fix, 0., and Livneh, Z. (1995) Deamination of cytosine- containing pyrimidine photodimers in UV-irradiated DNA. Significance for UV light mutagenesis. JBiol Chem 270, 24 174-24179.

46. Konrad, E. B., and Lehman, I. R. (1975) Novel mutants of Escherichia coli that accumulate very small DNA replicative intermediates. Proc Natl Acad Sci USA 72, 2150-2154.

47. Duncan, B. K., Rockstroh, P. A., and Warner, H. R. (1978) Escherichia coli K- 12 mutants deficient in uracil-DNA glycosylase. JBacteriol 134, 1039-1045. 239

48. Warner, H. R., and Duncan, B. K. (1978) In vivo synthesis and properties of uracil-containing DNA. Nature 272, 32-34.

49. Fisher, E. F., and Caruthers, M.H. (1979) Studies on gene control regons XII. The functional significance of a lac operator constitutive mutation. Nucleic Acids Res. 7,401-416.

50. Verri, A., Mazzarello, P., Biamonti, G., Spadari, S., and Focher, F. (1990) The specific binding of nuclear protein(s) to the cAl'vlP responsive element (CRE) sequence (TGACGTCA) is reduced by the misincorporation of U and increased by the deamination of C. Nucleic Acids Res 18, 5775-5780.

51. Berkner, K. L., and Folk, W. R. (1979) The effects of substituted pyrimidines in DNAs on cleavage by sequence-specific endonucleases. JBiol Chem 254, 2551-2560.

52. Miller, J. H., Lebkowski, J. S., Greisen, K. S., and Cabs, M. P. (1984) Specificity of mutations induced in transfected DNA by mammalian cells. EMBOJ.3,3117-3121.

53. Yatagai, F., and Glickman, B. W. (1990) Specificity of spontaneous mutation in the lad gene cloned into bacteriophage M13. Mutat Res 243, 2 1-28.

54. de Jong, P. J., Grosovsky, A. J., and Glickman, B. W. (1988) Spectrum of spontaneous mutation at the APRT locus of Chinese hamster ovary cells: an analysis at the DNA sequence level. Proc NatiAcadSci USA 85, 3499-3503.

55. Impellizzeri, K. J., Anderson, B., and Burgers, P. M. (1991) The spectrum of spontaneous mutations in a Saccharomyces cerevisiae uracil-DNA-glycosylase mutant limits the function of this enzyme to cytosine deamination repair. J Bacteriol 173, 6807-68 10.

56. Lindahl, T., Ljungquist, S., Siegert, W., Nyberg, B., and Sperens, B. (1977) DNA N-glycosidases: properties of uracil-DNA glycosidase from Escherichia coil. JBioi Chem 252, 3286-3294.

57. Mosbaugh, D. W., and Bennett, S.E. (1994) Uracil-excision DNA repair. Prog. Nucleic Acid Res. Mol. Biol. 48, 315-370.

58. Pearl, L. H. (2000) Structure and function in the uracil-DNA glycosylase superfamily. Mutat. Res. 460, 165-191. 240

59. Aravind, L., and Koonin, E. V. (2000) The alphalbeta fold uracil DNA glycosylases: a common origin with diverse fates. Genome Biol 1, RESEARCH0007.

60. Varshney, U., Hutcheon, T., and van de Sande, J. H. (1988) Sequence analysis, expression, and conservation of Escherichia co/i uracil DNA glycosylase and its gene (ung). JBiol Chem 263, 7776-7784.

61. Duncan, B. K., and Chambers, J. A. (1984) The cloning and overproduction of Escherichia co/i uracil-DNA glycosylase. Gene 28, 211-219.

62. Bennett, S. E., Sanderson, R. J., and Mosbaugh, D. W. (1995) Processivity of Escherichia co/i and rat liver mitochondrial uracil-DNA glycosylase is affected by NaC1 concentration. Biochemistry 34, 6109-6119.

63. Kumar, N. V., and Varshney, U. (1997) Contrasting effects of single stranded DNA binding protein on the activity of uracil DNA glycosylase from Escherichia co/i towards different DNA substrates. Nucleic Acids Res 25, 2336-2343.

64. Ghosh, M., Rumpal, N., Varshney, U., and Chary, K. V. (2002) Structural basis for poor uracil excision from hairpin DNA. An NMR study. Eur J Biochem 269, 1886-1894.

65. Mauro, D. J., De Riel, J. K., Tallarida, R. J., and Sirover, M. A. (1993) Mechanisms of excision of 5-fluorouracil by uracil DNA glycosylase in normal human cells. Mol. Pharmacol. 43, 854-857.

66. Hatahet, Z., Kow, Y. W., Purmal, A. A., Cunningham, R. P., and Wallace, S. 5. (1994) New substrates for old enzymes. 5-hydroxy-2'-deoxycytidine and 5- hydroxy-2'-deoxyuridine are substrates for Escherichia co/i endonuclease III and formamidopyrimidine DNA N-glycosylase, while 5-hydroxy-2'- deoxyuridine is a substrate for uracil DNA N-glycosylase. J. Biol. Chem. 269, 188 14-18820.

67. Zastawny, T. H., Doetsch, P. W., and Dizdaroglu, M. (1995) A novel activity of E. coli DNA N-glycosylase. Excision of isodialuric acid (5,6- dihydroxyuracil), a major product of oxidative DNA damage, from DNA. FEBSLett. 364,255-258.

68. Varshney, U., and van de Sande, J. H. (1991) Specificities and kinetics of uracil excision from uracil-containing DNA oligomers by Escherichia co/i uracil DNA glycosylase. Biochemistry 30,4055-4061. 241

69. Purmal, A. A., Wallace, S. S., and Kow, Y. W. (1996) The phosphodiester bond 3' to a deoxyuridine residue is crucial for substrate binding for uracil DNA N-glycosylase. Biochemistry 35, 16630-16637.

70. Jiang, Y. L., Ichikawa, Y., Song, F., and Stivers, J. T. (2003) Powering DNA repair through substrate electrostatic interactions. Biochemistry 42, 1922-1929.

71. Nilsen, H., Yazdankhah, S. P., Eftedal, I., and Krokan, H. E. (1995) Sequence specificity for removal of uracil from U:A pairs and U:G mismatches by uracil-DNA glycosylase from Escherichia coli, and correlation with mutational hotspots. FEBSLett. 362, 205-209.

72. Berg, 0. G., Winter, R. B., and von Hippel, P. H. (1981) Diffusion-driven mechanisms of protein translocation on nucleic acids. 1. Models and theory. Biochemistry 20, 6929-6948.

73. Higley, M., and Lloyd, R. S. (1993) Processivity of uracil DNA glycosylase. MutatRes 294, 109-116.

74. Purmal, A. A., Lampman, G. W., Pourmal, E. I., Melamede, R. J., Wallace, S. S., and Kow, Y. W. (1994) Uracil DNA N-glycosylase distributively interacts with duplex polynucleotides containing repeating units of either TGGCCAAGCU or TGGCCAAGCTTGGCCAAGCU. JBiol Chem 269, 22046-22053.

75. Gallinari, P., and Jiricny, J. (1996) A new class of uracil-DNA glycosylases related to human thymine-DNA glycosylase. Nature 383, 735-73 8.

76. Barrett, T. E., Savva, R., Panayotou, G., Barlow, T., Brown, T., Jiricny, J., and Pearl, L. H. (1998) Crystal structure of a G:T/U mismatch-specific DNA glycosylase: mismatch recognition by complementary-strand interactions. Cell 92, 117-129.

77. Saparbaev, M., and Laval, J. (1998) 3,N4-ethenocytosine, a highly mutagenic adduct, is a primary substrate for Escherichia coli double-stranded uracil-DNA glycosylase and human mismatch-specific thymine-DNA glycosylase. Proc NatlAcadSci USA 95, 8508-8513.

78. Sung, J. S., and Mosbaugh, D. W. (2000) Escherichia coli double-strand uracil-DNA glycosylase: involvement in uracil-mediated DNA base excision repair and stimulation of activity by endonuclease IV. Biochemistry 39, 10224- 1023 5. 242

79. Lutsenko, E., and Bhagwat, A. S. (1999) The role of the Escherichia co/i mug protein in the removal of uracil and 3,N(4)-ethenocytosine from DNA. JBiol Chem 274, 31034-31038.

80. Hennecke, F., Kolmar, H., Brundi, K., and Fritz, H. J. (1991) The vsr gene product of E. coli K-12 is a strand- and sequence-specific DNA mismatch endonuclease. Nature 353, 776-778.

81. Sohail, A., Lieb, M., Dar, M., and Bhagwat, A. S. (1990) A gene required for very short patch repair in Escherichia co/i is adjacent to the DNA cytosine methylase gene. JBacteriol 172,4214-4221.

82. Sung, J. S., Bennett, S. E., and Mosbaugh, D. W. (2001) Fidelity of uracil- initiated base excision DNA repair in Escherichia coli cell extracts. JBiol Chem 276, 2276-2285. Epub 2000 Oct 2216.

83. Barrett, T. E., Scharer, 0. D., Savva, R., Brown, T., Jiricny, J., Verdine, G. L., and Pearl, L. H. (1999) Crystal structure of a thwarted mismatch glycosylase DNA repair complex. Embo J 18, 6599-6609.

84. Saparbaev, M., Langouet, S., Privezentzev, C. V., Guengerich, F. P., Cai, H., Elder, R. H., and Laval, J. (2002) 1,N(2)-ethenoguanine, a mutagenic DNA adduct, is a primary substrate of Escherichia co/i mismatch-specific uracil- DNA glycosylase and human alkylpurine-DNA-N-glycosylase. JBiol Chem 277, 26987-26993.

85. Hang, B., Downing, G., Guliaev, A. B., and Singer, B. (2002) Novel activity of Escherichia co/i mismatch uracil-DNA glycosylase (Mug) excising 8- (hydroxymethyl)-3,N4-ethenocytosine, a potential product resulting from glycidaldehyde reaction. Biochemistry 41, 2158-2165.

86. Domena, J. D., and Mosbaugh, D. W. (1985) Purification of nuclear and mitochondrial uracil-DNA glycosylase from rat liver. Identification of two distinct subcellular forms. Biochemistry 24, 7320-7328.

87. Domena, J. D., Timmer, R. T., Dicharry, S. A., and Mosbaugh, D. W. (1988) Purification and properties of mitochondrial uracil-DNA glycosylase from rat liver. Biochemistry 27, 6742-6751.

88. Talpaert-Borle, M., Clerici, L., and Campagnari, F. (1979) Isolation and characterization of a uracil-DNA glycosylase from calf thymus. JBiol Chem 254, 6387-6391. 243

89. Talpaert-Borle, M., Campagnari, F., and Creissen, D. M. (1982) Properties of purified uracil-DNA glycosylase from calf thymus. An in vitro study using synthetic DNA-like substrates. JBiol Chem 257, 1208-1214.

90. Caradonna, S. J., and Cheng, Y. C. (1980) Uracil DNA-glycosylase. Purification and properties of this enzyme isolated from blast cells of acute myelocytic leukemia patients. JBiol Chem 255, 2293-2300.

91. Wittwer, C. U., and Krokan, H. (1985) Uracil-DNA glycosylase in HeLa S3 cells: interconvertibility of 50 and 20 kDa forms and similarity of the nuclear and mitochondrial form of the enzyme. Biochim Biophys Acta 832, 308-318.

92. Wittwer, C. U., Bauw, G., and Krokan, H. E. (1989) Purification and determination of the NH2-terminal amino acid sequence of uracil-DNA glycosylase from human placenta. Biochemistry 28, 780-784.

93. Leblanc, J. P., and Laval, J. (1982) Comparison at the molecular level of uracil-DNA glycosylases from different origins. Biochimie 64, 735-738.

94. Gupta, P. K., and Sirover, M. A. (1981) Stimulation of the nuclear uracil DNA glycosylase in proliferating human fibroblasts. Cancer Res 41, 3133-3136.

95. Eftedal, I., Guddal, P. H., Slupphaug, G., Volden, G., and Krokan, H. E. (1993) Consensus sequences for good and poor removal of uracil from double stranded DNA by uracil-DNA glycosylase. Nucleic Acids Res 21,2095-2101.

96. Olsen, L. C., Aasland, R., Krokan, H. E., and Helland, D. E. (1991) Human uracil-DNA glycosylase complements E. coli ung mutants. Nucleic Acids Res 19,4473-4478.

97. Anderson, C. T., and Friedberg, E. C. (1980) The presence of nuclear and mitochondrial uracil-DNA glycosylase in extracts of human KB cells. Nucleic Acids Res 8, 875-888.

98. Krokan, H., and Wittwer, C. U. (1981) Uracil DNA-glycosylase from HeLa cells: general properties, substrate specificity and effect of uracil analogs. Nucleic Acids Res 9, 2599-26 13.

99. Slupphaug, G., Markussen, F. H., Olsen, L. C., Aasland, R., Aarsaether, N., Bakke, 0., Krokan, H. E., and Helland, D. E. (1993) Nuclear and mitochondrial forms of human uracil-DNA glycosylase are encoded by the same gene. Nucleic Acids Res 21, 2579-2584. 244

100. Slupphaug, G., Eftedal, I., Kavli, B., Bharati, S., Helle, N. M., Haug, T., Levine, D. W., and Krokan, H. E. (1995) Properties of a recombinant human uracil-DNA glycosylase from the UNG gene and evidence that UNG encodes the major uracil-DNA glycosylase. Biochemistry 34, 128-13 8.

101. Nagelhus, T. A., Slupphaug, G., Lindmo, T., and Krokan, H. E. (1995) Cell cycle regulation and subcellular localization of the major human uracil-DNA glycosylase. Exp Cell Res 220, 292-297.

102. Haug, T., Skorpen, F., Kvaloy, K., Eftedal, I., Lund, H., and Krokan, H. E. (1996) Human uracil-DNA glycosylase gene: sequence organization, methylation pattern, and mapping to chromosome 12q23-q24.1. Genomics 36, 408-4 16.

103. Haug, T., Skorpen, F., Aas, P. A., Malm, V., Skjelbred, C., and Krokan, H. E. (1998) Regulation of expression of nuclear and mitochondrial forms of human uracil-DNA glycosylase. Nucleic Acids Res 26, 1449-1457.

104. Nilsen, H., Otterlei, M., Haug, T., Solum, K., Nageihus, T. A., Skorpen, F., and Krokan, H. E. (1997) Nuclear and mitochondrial uracil-DNA glycosylases are generated by alternative splicing and transcription from different positions in the UNG gene. Nucleic Acids Res 25, 750-755.

105. Otterlei, M., Haug, T., Nagelhus, T. A., Slupphaug, G., Lindmo, T., and Krokan, H. E. (1998) Nuclear and mitochondrial splice forms of human uracil- DNA glycosylase contain a complex nuclear localisation signal and a strong classical mitochondrial localisation signal, respectively. Nucleic Acids Res 26, 4611-4617.

106. Otterlei, M., Warbrick, E., Nagelhus, T. A., Haug, T., Slupphaug, G., Akbari, M., Aas, P. A., Steinsbekk, K., Bakke, 0., and Krokan, H. E. (1999) Post- replicative base excision repair in replication foci. Embo J 18, 3834-3844.

107. Krokan, H. E., Otterlei, M., Nilsen, H., Kavli, B., Skorpen, F., Andersen, S., Skjelbred, C., Akbari, M., Aas, P. A., and Slupphaug, G. (2001) Properties and functions of human uracil-DNA glycosylase from the UNG gene. Frog Nucleic Acid Res Mol Biol 68, 365-3 86.

108. Muller-Weeks, S., Mastran, B., and Caradonna, S. (1998) The nuclear isoform of the highly conserved human uracil-DNA glycosylase is an Mr 36,000 phosphoprotein. JBiol Chem 273,21909-21917. 245

109. Dizdaroglu, M., Zastawny, T. H., Carmical, J. R., and Lloyd, R. S. (1996) A novel DNA N-glycosylase activity of E. coli T4 endonuclease V that excises 4,6-diamino-5-formamidopyrimidine from DNA, a UV-radiation- and hydroxyl radical-induced product of adenine. MutatRes 362, 1-8.

110. Savva, R., McAuley-Hecht, K., Brown, T., and Pearl, L. (1995) The structural basis of specific base-excision repair by uracil-DNA glycosylase. Nature 373, 487-493.

111. Mol, C. D., Arvai, A. S., Slupphaug, G., Kavli, B., Alseth, I., Krokan, H. E., and Tam, J. A. (1995) Crystal structure and mutational analysis of human uracil-DNA glycosylase: structural basis for specificity and catalysis. Cell 80, 869-878.

112. Xiao, G., Tordova, M., Jagadeesh, J., Drohat, A. C., Stivers, J. T., and Gilliland, G. L. (1999) Crystal structure of Escherichia coli uracil DNA glycosylase and its complexes with uracil and glycerol: structure and glycosylase mechanism revisited. Proteins 35, 13-24.

113. Parikh, S. S., Putnam, C. D., and Tam, J. A. (2000) Lessons learned from structural results on uracil-DNA glycosylase. Mutat Res 460, 183-199.

114. Putnam, C. D., Shroyer, M. J., Lundquist, A. J., Mol, C. D., Arvai, A. S., Mosbaugh, D. W., and Tamer, J. A. (1999) Protein mimicry of DNA from crystal structures of the uracil-DNA glycosylase inhibitor protein and its complex with Escherichia coli uracil-DNA glycosylase. JMolBiol 287, 331- 346.

115. Slupphaug, G., Mo!, C. D., Kavli, B., Arvai, A. S., Krokan, H. E., and Tamer, J. A. (1996) A nucleotide-flipping mechanism from the structure of human uracil-DNA glycosylase bound to DNA. Nature 384, 87-92.

116. Parikh, S. S., Mo!, C. D., S!upphaug, G., Bharati, S., Krokan, H. B., and Tamer, J. A. (1998) Base excision repair initiation revealed bycrystal structures and binding kinetics of human uracil-DNA glycosylase with DNA. Embo J 17, 52 14-5226.

117. Parikh, S. S., Waicher, G., Jones, G. D., Slupphaug, G., Krokan, H. E., Blackburn, G. M., and Tam, J. A. (2000) Uracil-DNA glycosy!ase-DNA substrate and product structures: conformational strain promotes catalytic efficiency by coupled stereoe!ectronic effects. Proc NatlAcadSci USA 97, 5083-5088. LI

118. Stivers, J. T., Pankiewicz, K. W., and Watanabe, K. A. (1999) Kinetic mechanism of damage site recognition and uracil flipping by Escherichia coil uracil DNA glycosylase. Biochemistry 38, 952-963.

119. Drohat, A. C., Jagadeesh, J., Ferguson, E., and Stivers, J. 1. (1999) Role of electrophilic and general base catalysis in the mechanism of Escherichia coil uracil DNA glycosylase. Biochemistry 38, 11866-11875.

120. Jiang, Y. L., and Stivers, J. 1. (2002) Mutational analysis of the base-flipping mechanism of uracil DNA glycosylase. Biochemistry 41, 11236-11247.

121. Wong, I., Lundquist, A. J., Bernards, A. S., and Mosbaugh, D. W. (2002) Presteady-state analysis of a single catalytic turnover by Escherichia coli uracil-DNA glycosylase reveals a "pinch-pull-push" mechanism. JBiol Chem 277, 19424-19432.

122. Jiang, Y. L., Stivers, J. T., and Song, F. (2002) Base-flipping mutations of uracil DNA glycosylase: substrate rescue using a pyrene nucleotide wedge. Biochemisfty 41, 11248-11254.

123. Kavli, B., Slupphaug, G., Mol, C. D., Arvai, A. S., Peterson, S. B., Tamer, J. A., and Krokan, H. E. (1996) Excision of cytosine and thymine from DNA by mutants of human uracil-DNA glycosylase. Embo J15, 3442-3447.

124. Shroyer, M. J., Bennett, S. E., Putnam, C. D., Tamer, J. A., and Mosbaugh, D. W. (1999) Mutation of an active site residue in Escherichia coil uracil-DNA glycosylase: effect on DNA binding, uracil inhibition and catalysis. Biochemistry 38, 4834-4845.

125. Werner, R. M., and Stivers, J. T. (2000) Kinetic isotope effect studies of the reaction catalyzed by uracil DNA glycosylase: evidence for an oxocarbenium ion-uracil anion intermediate. Biochemistry 39, 14054-14064.

126. Jiang, Y. L., Drohat, A. C., Ichikawa, Y., and Stivers, J. T. (2002) Probing the limits of electrostatic catalysis by uracil DNA glycosylase using transition state mimicry and mutagenesis. JBiol Chem 277, 15385-15392.

127. Jiang, Y. L., Cao, C., Stivers, J. T., Song, F., and Ichikawa, Y. (2004) The merits of bipartite transition-state mimics for inhibition of uracil DNA glycosylase. Bioorg Chem 32,244-262.

128. Dinner, A. R., Blackburn, G. M., and Karplus, M. (2001) Uracil-DNA glycosylase acts by substrate autocatalysis. Nature 413, 752-755. 247

129. Wiebauer, K., and Jiricny, J. (1989) In vitro correction of G.T mispairs to G.0 pairs in nuclear extracts from human cells. Nature 339, 234-236.

130. Wiebauer, K., and Jiricny, J. (1990) Mismatch-specific thymine DNA glycosylase and DNA polymerase beta mediate the correction of G.T mispairs in nuclear extracts from human cells. Proc NatlAcadSci USA 87, 5842-5845.

131. Neddermann, P., and Jiricny, J. (1993) The purification of a mismatch-specific thymine-DNA glycosylase from HeLa cells. JBiol Chem 268, 21218-21224.

132. Neddermann, P., Gallinari, P., Lettieri, T., Schmid, D., Truong, 0., Hsuan, J. J., Wiebauer, K., and Jiricny, J. (1996) Cloning and expression of human G/T mismatch-specific thymine-DNA glycosylase. JBiol Chem 271, 12767-12774.

133. Sard, L., Tomielli, S., Gallinari, P., Minoletti, F., Jiricny, J., Lettieri, T., Pierotti, M. A., Sozzi, G., and Radice, P. (1997) Chromosomal localizations and molecular analysis of TDG gene-related sequences. Genomics 44, 222- 226. 134. Waters, T. R., and Swami, P. F. (1998) Kinetics of the action of thymine DNA glycosylase. JBiol Chem 273, 20007-20014.

135. Sibghat, U., and Day, R. S., 3rd. (1992) Incision at 06-methylguanine:thymine mispairs in DNA by extracts of human cells. Biochemistry 31, 7998-8008.

136. Sibghat, U., and Day, R. S., 3rd. (1993) DNA-substrate sequence specificity of human G:T mismatch repair activity. Nucleic Acids Res 21, 128 1-1287.

137. Griffm, S., and Karran, P. (1993) Incision at DNA G.T mispairs by extracts of mammalian cells occurs preferentially at cytosine methylation sites and is not targeted by a separate G.T binding reaction. Biochemistry 32, 13032-13039.

138. Griffin, S., Branch, P., Xu, Y. Z., and Karran, P. (1994) DNA mismatch binding and incision at modified guanine bases by extracts of mammalian cells: implications for tolerance to DNA methylation damage. Biochemistry 33, 4787-4793.

139. Sibghat, U., Gallinari, P., Xu, Y. Z., Goodman, M. F., Bloom, L. B., Jiricny, J., and Day, R. S., 3rd. (1996) Base analog and neighboring base effects on substrate specificity of recombinant human G:T mismatch-specific thymine DNA-glycosylase. Biochemistry 35, 12926-12932.

140. Waters, T. R., Gallinari, P., Jiricny, J., and Swami, P. F. (1999) Human thymine DNA glycosylase binds to apurinic sites in DNA but is displaced by human apurinic endonuclease 1. JBiol Chem 274, 67-74. 248

141. Scharer, 0. D., Nash, H. M., Jiricny, J., Laval, J., and Verdine, G. L. (1998) Specific binding of a designed pyrrolidine abasic site analog to multiple DNA glycosylases. JBiol Chem 273, 8592-8597.

142. Hardeland, U., Steinacher, R., Jiricny, J., and Schar, P. (2002) Modification of the human thymine-DNA glycosylase by ubiquitin-like proteins facilitates enzymatic turnover. Embo J21, 1456-1464.

143. Yoon, J. H., Iwai, S., O'Connor, T. R., and Pfeifer, G. P. (2003) Human thymine DNA glycosylase (TDG) and methyl-CpG-binding protein 4 (MBD4) excise thymine glycol (Tg) from a Tg:G mispair. Nucleic Acids Res 31, 5399- 5404.

144. Hang, B., Medina, M., Fraenkel-Conrat, H., and Singer, B. (1998) A 55-kDa protein isolated from human cells shows DNA glycosylase activity toward 3,N4-ethenocytosine and the GIlT mismatch. Proc NatlAcadSci USA 95, 13561-13566.

145. Urn, S., Harbers, M., Benecke, A., Pierrat, B., Losson, R., and Chambon, P. (1998) Retinoic acid receptors interact physically and fl.rnctionally with the T:G mismatch-specific thymine-DNA glycosylase. JBiol Chem 273,20728- 20736.

146. Tini, M., Benecke, A., Urn, S. J., Torchia, J., Evans, R. M., and Chambon, P. (2002) Association of CBP/p300 acetylase and thymine DNA glycosylase links DNA repair and transcription. Mol Cell 9, 265-277.

147. Haushalter, K. A., Todd Stukenberg, M. W., Kirscbner, M. W., and Verdine, G. L. (1999) Identification of a new uracil-DNA glycosylase family by expression cloning using synthetic inhibitors. Curr Biol 9, 174-185.

148. Elateri, I., Muller-Weeks, S., and Caradonna, S. (2003) The transcription factor, NFI/CTF plays a positive regulatory role in expression of the hSMUG1 gene. DNA Repair (Amst) 2, 1371-1385.

149. Kavli, B., Sundheirn, 0., Akbari, M., Otterlei, M., Nilsen, H., Skorpen, F., Aas, P. A., Hagen, L., Krokan, H. E., and Slupphaug, G. (2002) hUNG2 is the major repair enzyme for removal of uracil from U:A matches, U:G mismatches, and U in single-stranded DNA, with hSMUG1 as a broad specificity backup. JBiol Chem 277, 39926-39936.

150. Elateri, I., Tinkelenberg, B. A., Hansbury, M., Caradonna, S., Muller-Weeks, S., and Ladner, R. D. (2003) hSMUG1 can functionally compensate for Ungi in the yeast Saccharomyces cerevisiae. DNA Repair (Amst) 2, 3 15-323. 249

151. Boorstein, R. J., Cummings, A., Jr., Marenstein, D. R., Chan, M. K., Ma, Y., Neubert, T. A., Brown, S. M., and Teebor, G. W. (2001) Defmitive identification of mammalian 5-hydroxymethyluracil DNA N-glycosylase activity as SMUG1. JBiol Chem 276,41991-41997.

152. Masaoka, A., Matsubara, M., Hasegawa, R., Tanaka, T., Kurisu, S., Terato, H., Ohyama, Y., Karino, N., Matsuda, A., and Ide, H. (2003) Mammalian 5- formyluracil-DNA glycosylase. 2. Role of SMUG 1 uracil-DNA glycosylase in repair of 5-formyluracil and other oxidized and deaminated base lesions. Biochemistry 42, 5003-50 12.

153. Bird, A. (1992) The essentials of DNA methylation. Cell 70, 5-8.

154. Singal, R., and Ginder, G. D. (1999) DNA methylation. Blood 93, 4059-4070.

155. Wade, P. A. (2001) Methyl CpG-binding proteins and transcriptional repression. Bioessays 23, 1131-1137.

156. Hendrich, B., and Bird, A. (1998) Identification and characterization of a family of mammalian methyl-CpG binding proteins. Mo! Cell Biol 18, 6538- 6547.

157. Bellacosa, A., Cicchillitti, L., Schepis, F., Riccio, A., Yeung, A. T., Matsumoto, Y., Golemis, E. A., Genuardi, M., and Neri, G. (1999) MED 1, a novel human methyl-CpG-binding endonuclease, interacts with DNA mismatch repair protein MLH1. Proc NatlAcadSci USA 96, 3969-3974.

158. Millar, C. B., Guy, J., Sansom, 0. J., Selfridge, J., MacDougall, B., Hendrich, B., Keightley, P. D., Bishop, S. M., Clarke, A. R., and Bird, A. (2002) Enhanced CpG mutability and tumorigenesis in MBD4-deficient mice. Science 297,403-405.

159. Petronzelli, F., Riccio, A., Markham, G. D., Seehoizer, S. H., Stoerker, J., Genuardi, M., Yeung, A. T., Matsumoto, Y., and Bellacosa, A. (2000) Biphasic kinetics of the human DNA repair protein MED 1 (MBD4), a mismatch-specific DNA N-glycosylase. JBiol Chem 275, 32422-32429.

160. Petronzelli, F., Riccio, A., Markham, G. D., Seeholzer, S. H., Genuardi, M., Karbowski, M., Yeung, A. T., Matsumoto, Y., and Bellacosa, A. (2000) Investigation of the substrate spectrum of the human mismatch-specific DNA N-glycosylase MED1 (MBD4): fundamental role of the catalytic domain. J Cell Physiol 185,473-480. 250

161. Cortellino, S., Turner, D., Masciullo, V., Schepis, F., Albino, D., Daniel, R., Skalka, A. M., Meropol, N. J., Alberti, C., Larue, L., and Bellacosa, A. (2003) The base excision repair enzyme MED1 mediates DNA damage response to antitumor drugs and is associated with mismatch repair system integrity. Proc NatlAcadSci USA 100, 15071-15076.

162. Takahashi, I., and Marmur, J. (1963) Replacement of thymidylic acid by deoxyuridylic acid in the deoxyribonucleic acid of a transducing phage for Bacillus subtilis. Nature 197, 794-795.

163. Price, A. R., and Fogt, S. M. (1973) Deoxythymidylate phosphohydrolase induced by bacteriophage PBS2 during infection of Bacillus subtilis. JBiol Chem 248, 1372-1380.

164. Tomita, F., and Takahashi, I. (1969) A novel enzyme, dCTP deaminase, found in Bacillus subtilis infected with phage PBS I. Biochim Biophys Acta 179, 1 8- 27.

165. Kahan, F. M. (1963) Novel enzymes formed by Bacillus subtilis infected with bacteriophage. Federation Proceedings 22,406.

166. Price, A. R., and Frato, J. (1975) Bacillus subtilis deoxyuridinetriphosphatase and its bacteriophage PBS2-induced inhibitor. JBiol Chem 250, 8804-8811.

167. Katz, G. E., Price, A. R., and Pomerantz, M. J. (1976) Bacteriophage PBS2- induced inhibition of uracil-containing DNA degradation. J Virol 20, 535-538.

168. Friedberg, E. C., Ganesan, A. K., and Minton, K. (1975) N-Glycosidase activity in extracts of Bacillus subtilis and its inhibition after infection with bacteriophage PBS2. J Virol 16, 315-321.

169. Cone, R., Bonura, T., and Friedberg, E. C. (1980) Inhibitor of uracil-DNA glycosylase induced by bacteriophage PBS2. Purification and preliminary characterization. JBiol Chem 255, 10354-10358.

170. Wang, Z., and Mosbaugh, D. W. (1989) Uracil-DNA glycosylase inhibitor gene of bacteriophage PBS2 encodes a binding protein specific for uracil-DNA glycosylase. JBiol Chem 264, 1163-1171.

171. Bennett, S. E., and Mosbaugh, D. W. (1992) Characterization of the Escherichia coli uracil-DNA glycosylase.inhibitor protein complex. JBiol Chem 267, 22512-22521. MU

172. Wang, Z., and Mosbaugh, D. W. (1988) Uracil-DNA glycosylase inhibitor of bacteriophage PBS2: cloning and effects of expression of the inhibitor gene in Escherichia coli. JBacteriol 170, 1082-1091.

173. Savva, R., and Pearl, L. H. (1995) Nucleotide mimicry in the crystal structure of the uracil-DNA glycosylase-uracil glycosylase inhibitor protein complex. Nat Struct Biol 2, 752-757.

174. Williams, M. V., and Pollack, J. D. (1990) A mollicute (mycoplasma) DNA repair enzyme: purification and characterization of uracil-DNA glycosylase. J Bacteriol 172, 2979-2985.

175. Winters, T. A., and Williams, M. V. (1990) Use of the PBS2 uracil-DNA glycosylase inhibitor to differentiate the uracil-DNA glycosylase activities encoded by herpes simplex virus types 1 and 2. J Virol Methods 29, 233-242.

176. Radany, B. H., Domfeld, K. J., Sanderson, R. J., Savage, M. K., Majumdar, A., Seidman, M. M., and Mosbaugh, D. W. (2000) Increased spontaneous mutation frequency in human cells expressing the phage PBS2-encoded inhibitor of uracil-DNA glycosylase. MutatRes 461, 41-58.

177. Wang, Z. G., Smith, D. G., and Mosbaugh, D. W. (1991) Overproduction and characterization of the uracil-DNA glycosylase inhibitor of bacteriophage PBS2. Gene 99, 3 1-37.

178. Bennett, S. E., Schimerlik, M. I., and Mosbaugh, D. W. (1993) Kinetics of the uracil-DNA glycosylase/inhibitor protein association. Ung interaction with Ugi, nucleic acids, and uracil compounds. JBiol Chem 268, 26879-26885.

179. Bennett, S. B., Jensen, 0. N., Barofsky, D. F., and Mosbaugh, D. W. (1994) UV-catalyzed cross-linking of Escherichia co/i uracil-DNA glycosylase to DNA. Identification of amino acid residues in the single-stranded DNA binding site. JBiol Chem 269, 2 1870-21879.

180. Sanderson, R. J., and Mosbaugh, D. W. (1996) Identification of specific carboxyl groups on uracil-DNA glycosylase inhibitor protein that are required for activity. JBioI Chem 271, 29170-29181.

181. Lundquist, A. J., Beger, R. D., Bennett, S. B., Bolton, P. H., and Mosbaugh, D. W. (1997) Site-directed mutagenesis and characterization of uracil-DNA glycosylase inhibitor protein. Role of specific carboxylic amino acids in complex formation with Escherichia co/i uracil-DNA glycosylase. JBiol Chem 272,21408-21419. 252

182. Balasubramanian, S., Beger, R. D., Bennett, S. B., Mosbaugh, D. W., and Bolton, P. H. (1995) Secondary structure of uracil-DNA glycosylase inhibitor protein. JBiol Chem 270, 296-303.

183. Mol, C. D., Arvai, A. S., Sanderson, R. J., Slupphaug, G., Kavli, B., Krokan, H. B., Mosbaugh, D. W., and Tam, J. A. (1995) Crystal structure of human uracil-DNA glycosylase in complex with a protein inhibitor: protein mimicry of DNA. Cell 82, 701-708.

184. Beger, R. D., Balasubramanian, S., Bennett, S. E., Mosbaugh, D. W., and Bolton, P. H. (1995) Tertiary structure of uracil-DNA glycosylase inhibitor protein. JBiol Chem 270, 16840-16847.

185. Krokan, H. B., Standal, R., and Slupphaug, G. (1997) DNA glycosylases in the base excision repair of DNA. Biochem J325, 1-16.

186. Demple, B., Herman, T., and Chen, D. S. (1991) Cloning and expression of APE, the cDNA encoding the major human apurinic endonuclease: definition of a family of DNA repair enzymes. Proc NatlAcadSci USA 88, 11450- 11454.

187. Xanthoudakis, S., and Curran, T. (1992) Identification and characterization of Ref-i, a nuclear protein that facilitates AP- 1 DNA-binding activity. Embo J 11, 653-665.

188. Xanthoudakis, S., Miao, G., Wang, F., Pan, Y. C., and Curran, T. (1992) Redox activation of Fos-Jun DNA binding activity is mediated by a DNA repair enzyme. Embo J 11, 3323-3335.

189. Ramotar, D., Popoff, S.C., Gralla, E. B., and Demple, B. (1991) Cellular role of yeast Apni apurinic endonuclease/3 '-diesterase: repair of oxidative and alkylation DNA damage and control of spontaneous mutation. Mo! Cell Biol 11, 4537-4544.

190. Barzilay, G., Mo!, C. D., Robson, C. N., Walker, L. J., Cunningham, R. P., Tam, J. A., and Hickson, I. D. (1995) Identification of critical active-site residues in the multifunctional human DNA repair enzyme HAP 1. Nat Struct Biol 2, 561-568.

191. Mo!, C. D., Hosfield, D. J., and Tam, J. A. (2000) Abasic site recognition by two apurinic/apyrimidinic endonuclease families in DNA base excision repair: the 3' ends justify the means. MutatRes 460, 211-229. 253

192. Marenstein, D. R., Wilson, D. M., 3rd, and Teebor, G. W. (2004) Human AP endonuclease (APE 1) demonstrates endonucleolytic activity against AP sites in single-stranded DNA. DNA Repair (Amst) 3, 527-533.

193. Chen, D. S., Herman, T., and Demple, B. (1991) Two distinct human DNA diesterases that hydrolyze 3'-blocking deoxyribose fragments from oxidized DNA. Nucleic Acids Res 19, 5907-59 14.

194. Suh, D., Wilson, D. M., 3rd, and Povirk, L. F. (1997) 3'-phosphodiesterase activity of human apurinic/apyrimidinic endonuclease at DNA double-strand break ends. Nucleic Acids Res 25,2495-2500.

195. Winters, T. A., Henner, W. D., Russell, P. S., McCullough, A., and Jorgensen, T. J. (1994) Removal of 3'-phosphoglycolate from DNA strand-break damage in an oligonucleotide substrate by recombinant human apurinic/apyrimidinic endonuclease 1. Nucleic Acids Res 22, 1866-1873.

196. Wilson, D. M., 3rd, Takeshita, M., Grollman, A. P., and Demple, B. (1995) Incision activity of human apurinic endonuclease (Ape) at abasic site analogs in DNA. JBiol Chem 270, 16002-16007.

197. Wong, D., DeMott, M. S., and Demple, B. (2003) Modulation of the 3'-->5'- exonuclease activity of human apurinic endonuclease (Ape 1) by its 5'-incised Abasic DNA product. JBiol Chem 278, 36242-36249.

198. Chou, K. M., and Cheng, Y. C. (2002) An exonucleolytic activity of human apurinic/apyrimidinic endonuclease on 3' mispaired DNA. Nature 415, 655- 659.

199. Xanthoudakis, S., Smeyne, R. J., Wallace, J. D., and Curran, T. (1996) The redox/DNA repair protein, Ref-i, is essential for early embryonic development in mice. Proc NatlAcadSci USA 93, 89 19-8923.

200.Hadi, M. Z., and Wilson, D. M., 3rd. (2000) Second human protein with homology to the Escherichia coli abasic endonuclease exonuclease III. 312- 324.

201. Tsuchimoto, D., Sakai, Y., Sakumi, K., Nishioka, K., Sasaki, M., Fujiwara, T., and Nakabeppu, Y. (2001) Human APE2 protein is mostly localized in the nuclei and to some extent in the mitochondria, while nuclear APE2 is partly associated with proliferating cell nuclear antigen. Nucleic Acids Res 29, 2349- 2360. 254

202. Dianov, G., and Lindahl, T. (1994) Reconstitution of the DNA base excision- repair pathway. Curr Biol 4, 1069-1076.

203. Frosina, G., Fortini, P., Rossi, 0., Carrozzino, F., Raspaglio, G., Cox, L. S., Lane, D. P., Abbondandolo, A., and Dogliotti, E. (1996) Two pathways for base excision repair in mammalian cells. JBiol Chem 271, 9573-9578.

204. Kiungland, A., and Lindahi, T. (1997) Second pathway for completion of human DNA base excision-repair: reconstitution with purified proteins and requirement for DNase IV (FEN1). EMBO J 16, 3341-3348.

205. Singhal, R. K., Prasad, R., and Wilson, S. H. (1995) DNA polymerase beta conducts the gap-filling step in uracil-initiated base excision repair in a bovine testis nuclear extract. JBiol Chem 270, 949-957.

206. Kubota, Y., Nash, R. A., Kiungland, A., Schar, P., Barnes, D. E., and Lindahl, T. (1996) Reconstitution of DNA base excision-repair with purified human proteins: interaction between DNA polymerase beta and the XRCC1 protein. EMBO J 15, 6662-6670.

207. Nicholl, I. D., Nealon, K., and Kenny, M. K. (1997) Reconstitution of human base excision repair with purified proteins. Biochemistry 36, 7557-7566.

208. Nealon, K., Nicholl, I. D., and Kenny, M. K. (1996) Characterization of the DNA polymerase requirement of human base excision repair. Nucleic Acids Res 24, 3763-3770.

209. Weiser, T., Gassmann, M., Thommes, P., Ferrari, E., Hafkemeyer, P., and Hubscher, U. (1991) Biochemical and functional comparison of DNA polymerases alpha, delta, and epsilon from calf thymus. JBiol Chem 266, 10420- 10428.

210. Sobol, R. W., Horton, J. K., Kuhn, R., Gu, H., Singhal, R. K., Prasad, R., Rajewsky, K., and Wilson, S. H. (1996) Requirement of mammalian DNA polymerase-beta in base-excision repair. Nature 379, 183-186.

211. Matsumoto, Y., and Kim, K. (1995) Excision of deoxyribose phosphate residues by DNA polymerase beta during DNA repair. Science 269, 699-702.

212. Podlutsky, A. J., Dianova, II, Wilson, S. H., Bohr, V. A., and Dianov, G. L. (2001) DNA synthesis and dRPase activities of polymerase beta are both essential for single-nucleotide patch base excision repair in mammalian cell extracts. Biochemistry 40, 809-8 13. 255

213. Fortini, P., Pascucci, B., Parlanti, E., Sobol, R. W., Wilson, S. H., and Dogliotti, E. (1998) Different DNA polymerases are involved in the short- and long-patch base excision repair in mammalian cells. Biochemistry 37, 3 575- 3 580.

214. Stucki, M., Pascucci, B., Parlanti, E., Fortini, P., Wilson, S. H., Hubscher, U., and Dogliotti, E. (1998) Mammalian base excision repair by DNA polymerases delta and epsilon. Oncogene 17, 835-843.

215. Bennett, S. E., Sung, J. S., and Mosbaugh, D. W. (2001) Fidelity of uracil- initiated base excision DNA repair in DNA polymerase beta-proficient and- deficient mouse embryonic fibroblast cell extracts. JBiol Chem 276, 42588- 42600.

216. Parlanti, E., Pascucci, B., Terrados, G., Blanco, L., and Dogliotti, E. (2004) Aphidicolin-resistant and -sensitive base excision repair in wild-type and DNA polymerase beta-defective mouse cells. DNA Repair (Amst) 3, 703-7 10.

217.Bebenek, K., Tissier, A., Frank, E. G., McDonald, J. P., Prasad, R., Wilson, S. H., Woodgate, R., and Kunkel, T. A. (2001) 5'-Deoxyribose phosphate lyase activity of human DNA polymerase iota in vitro. Science 291, 2156-2159.

218. Garcia-Diaz, M., Bebenek, K., Kunkel, T. A., and Blanco, L. (2001) Identification of an intrinsic 5'-deoxyribose-5-phosphate lyase activity in human DNA polymerase lambda: a possible role in base excision repair. JBiol Chem 276, 34659-34663.

219. Longley, M. J., Prasad, R., Srivastava, D. K., Wilson, S. H., and Copeland, W. C. (1998) Identification of 5'-deoxyribose phosphate lyase activity in human DNA polymerase gamma and its role in mitochondrial base excision repair in vitro. Proc NatlAcadSci USA 95, 12244-12248.

220. Pelletier, H., Sawaya, M. R., Wolfie, W., Wilson, S. H., and Kraut, J. (1996) Crystal structures of human DNA polymerase beta complexed with DNA: implications for catalytic mechanism, processivity, and fidelity. Biochemistry 35, 12742-12761.

221. Sobol, R. W., Prasad, R., Evenski, A., Baker, A., Yang, X. P., Horton, J. K., and Wilson, S. H. (2000) The lyase activity of the DNA repair protein beta- polymerase protects from DNA-damage-induced cytotoxicity. Nature 405, 807-8 10. 256

222. Prasad, R., Beard, W. A., and Wilson, S. H. (1994) Studies of gapped DNA substrate binding by mammalian DNA polymerase beta. Dependence on 5'- phosphate group. JBi01 Chem 269, 18096-18101.

223. Sawaya, M. R., Prasad, R., Wilson, S. H., Kraut, J., and Pelletier, H. (1997) Crystal structures of human DNA polymerase beta complexed with gapped and nicked DNA: evidence for an induced fit mechanism. Biochemistiy 36, 11205- 112 15.

224. Osheroff, W. P., Jung, H. K., Beard, W. A., Wilson, S. H., and Kunkel, T. A. (1999) The fidelity of DNA polymerase beta during distributive and processive DNA synthesis. JBiol Chem 274, 3642-3650.

225. Srivastava, D. K., Berg, B. J., Prasad, R., Molina, J. T., Beard, W. A., Tomkinson, A. E., and Wilson, S. H. (1998) Mammalian abasic site base excision repair. Identification of the reaction sequence and rate-determining steps. JBIo1 Chem 273, 21203-21209.

226.Prasad, R., Dianov, G. L., Bohr, V. A., and Wilson, S. H. (2000) FEN1 stimulation of DNA polymerase beta mediates an excision step in mammalian long patch base excision repair. JBiol Chem 275, 4460-4466.

227. Dianov, G. L., Prasad, R., Wilson, S. H., and Bohr, V. A. (1999) Role of DNA polymerase beta in the excision step of long patch mammalian base excision repair. JBiol Chem 274, 13741-13743.

228. Matsumoto, Y., and Bogenhagen, D. F. (1989) Repair of a synthetic abasic site in DNA in a Xenopus laevis oocyte extract. Mo! Cell Biol 9, 3750-3757.

229. Matsumoto, Y., Kim, K., and Bogenhagen, D. F. (1994) Proliferating cell nuclear antigen-dependent abasic site repair in Xenopus laevis oocytes: an alternative pathway of base excision DNA repair. Mol Cell Biol 14, 6187- 6197.

230. Frosina, G., Fortini, P., Rossi, 0., Carrozzino, F., Abbondandolo, A., and Dogliotti, B. (1994) Repair of abasic sites by mammalian cell extracts. Biochem J304, 699-705.

231. Wang, Z., Wu, X., and Friedberg, E. C. (1997) Molecular mechanism of base excision repair of uracil-containing DNA in yeast cell-free extracts. JBiol Chem 272, 24064-2407 1. 257

232. Sanderson, R. J., Bennett, S. E., Sung, J. S., and Mosbaugh, D. W. Uracil- initiated base excision DNA repair synthesis fidelity in human colon adenocarcinoma LoVo and Escherichia coli cell extracts. 165-188.

233. Wang, Z., Wu, X., and Friedberg, E. C. (1993) DNA repair synthesis during base excision repair in vitro is catalyzed by DNA polymerase epsilon and is influenced by DNA polymerases alpha and delta in Saccharomyces cerevisiae. Mol Cell Biol 13, 1051-1058.

234.Kedar, P. S., Kim, S. J., Robertson, A., Hou, E., Prasad, R., Horton, J. K., and Wilson, S. H. (2002) Direct interaction between mammalian DNA polymerase beta and proliferating cell nuclear antigen. JBiol Chem 277, 31115-31123.

235. Kim, K., Biade, S., and Matsumoto, Y. (1998) Involvement of flap endonuclease 1 in base excision DNA repair. JBiol Chem 273, 8842-8848.

236. Gary, R., Kim, K., Cornelius, H. L., Park, M. S., and Matsumoto, Y. (1999) Proliferating cell nuclear antigen facilitates excision in long-patch base excision repair. JBiol Chem 274,4354-4363.

237. Tom, S., Ranalli, T. A., Podust, V. N., and Bambara, R. A. (2001) Regulatory roles of p21 and apurinic/apyrimidinic endonuclease 1 in base excision repair. JBiol Chem 276,48781-48789.

238. Warbrick, E. (1998) PCNA binding through a conserved motif. Bioessays 20, 195-199.

239.Maga, G., and Hubscher, U. (2003) Proliferating cell nuclear antigen (PCNA): a dancer with many partners. J Cell Sci 116, 305 1-3060.

240.Murante, R. S., Huang, L., Turchi, J. J., and Bambara, R. A. (1994) The calf 5'- to 3'-exonuclease is also an endonuclease with both activities dependent on primers annealed upstream of the point of cleavage. JBiol Chem 269, 1191- 1196.

241. Harrington, J. J., and Lieber, M. R. (1994) The characterization of a mammalian DNA structure-specific endonuclease. EMBO J 13, 1235-1246.

242. Liu, Y., Kao, H. I., and Bambara, R. A. Flap endonuclease 1: a central component of DNA metabolism. 589-6 15.

243. Ishimi, Y., Claude, A., Bullock, P., and Hurwitz, J. (1988) Complete enzymatic synthesis of DNA containing the SV4O origin of replication. JBiol Chem 263, 19723-19733. 258

244. Goulian, M., Richards, S. H., Heard, C. J., and Bigsby, B. M. (1990) Discontinuous DNA synthesis by purified mammalian proteins. JBiol Chem 265, 18461-18471.

245. Waga, S., Bauer, G., and Stiliman, B. (1994) Reconstitution of complete SV4O DNA replication with purified replication factors. JBiol Chem 269, 10923- 10934.

246. Tishkoff, D. X., Filosi, N., Gaida, G. M., and Kolodner, R. D. (1997) A novel mutation avoidance mechanism dependent on S. cerevisiae RAD27 is distinct from DNA mismatch repair. Cell 88,253-263.

247. Kucherlapati, M., Yang, K., Kuraguchi, M., Zhao, J., Lia, M., Heyer, J., Kane, M. F., Fan, K., Russell, R., Brown, A. M., Kneitz, B., Edelmann, W., Kolodner, R. D., Lipkin, M., and Kucherlapati, R. (2002) Haploinsufficiency of Flap endonuclease (Feni) leads to rapid tumor progression. Proc NatlAcad Sci USA 99, 9924-9929.

248. Shibata, Y., and Nakamura, T. (2002) Defective flap endonuclease 1 activity in mammalian cells is associated with impaired DNA repair and prolonged S phase delay. JBiol Chem 277, 746-754.

249. Dantzer, F., de La Rubia, G., Menissier-De Murcia, J., Hostomsky, Z., de Murcia, G., and Schreiber, V. (2000) Base excision repair is impaired in mammalian cells lacking Poly(ADP-ribose) polymerase-1. Biochemistry 39, 7559-7569.

250. Dantzer, F., Schreiber, V., Niedergang, C., Trucco, C., Flatter, E., De La Rubia, G., Oliver, J., Rolli, V., Menissier-de Murcia, J., and de Murcia, G. (1999) Involvement of poly(ADP-ribose) polymerase in base excision repair. Biochimie 81, 69-75.

251. Prasad, R., Lavrik, 0. I., Kim, S. J., Kedar, P., Yang, X. P., Vande Berg, B. J., and Wilson, S. H. (2001) DNA polymerase beta -mediated long patch base excision repair. Poly(ADP-ribose)polymerase- 1 stimulates strand displacement DNA synthesis. JBiol Chem 276, 32411-32414.

252. Cistulli, C., Lavrik, 0. I., Prasad, R., Hou, E., and Wilson, S. H. (2004) AP endonuclease and poly(ADP-ribose) polymerase- 1 interact with the same base excision repair intermediate. DNA Repair (Amst) 3, 581-591.

253.Bennett, R. A., Wilson, D. M., 3rd, Wong, D., and Demple, B. (1997) Interaction of human apurinic endonuclease and DNA polymerase beta in the base excision repair pathway. Proc NatlAcadSci USA 94, 7 166-7169. 259

254. Wong, D., and Demple, B. (2004) Modulation of the 5'-deoxyribose-5- phosphate lyase and DNA synthesis activities of mammalian DNA polymerase beta by apurinic/apyrimidinic endonuclease 1. JBiol Chem 279, 25268-25275.

255. Biswas, E. E., Zhu, F. X., and Biswas, S. B. (1997) Stimulation of RTH1 nuclease of the yeast Saccharomyces cerevisiae by replication protein A. Biochemistry 36, 5955-5962.

256.Nagelhus, T. A., Haug, T., Singh, K. K., Keshav, K. F., Skorpen, F., Otterlei, M., Bharati, S., Lindmo, T., Benichou, S., Benarous, R., and Krokan, H. E. (1997) A sequence in the N-terminal region of human uracil-DNA glycosylase with homology to XPA interacts with the C-terminal part of the 34-kDa subunit of replication protein A. JBiol Chem 272, 656 1-6566.

257. Wold, M. S. Replication protein A: a heterotrimeric, single-stranded DNA- binding protein required for eukaryotic DNA metabolism. 6 1-92.

258. Araujo, S. J., Tirode, F., Coin, F., Pospiech, H., Syvaoja, J. B., Stucki, M., Hubscher, U., Egly, J. M., and Wood, R. D. (2000) Nucleotide excision repair of DNA with recombinant human proteins: definition of the minimal set of factors, active forms of TFIIH, and modulation by CAK. Genes Dev. 14, 349- 359.

259.Mer, G., Bochkarev, A., Gupta, R., Bochkareva, B., Frappier, L., Ingles, C. J., Edwards, A. M., and Chazin, W. J. (2000) Structural basis for the recognition of DNA repair proteins UNG2, XPA, and RAD52 by replication factor RPA. Cell 103,449-456.

260. DeMott, M. S., Zigman, S., and Bambara, R. A. (1998) Replication protein A stimulates long patch DNA base excision repair. JBiol Chem 273, 27492- 27498.

261. Dianov, G. L., Jensen, B. R., Kenny, M. K., and Bohr, V. A. (1999) Replication protein A stimulates proliferating cell nuclear antigen-dependent repair of abasic sites in DNA by human cell extracts. Biochemistry 38, 11021- 11025.

262. Ranalli, T. A., DeMott, M. S., and Bambara, R. A. (2002) Mechanism underlying replication protein a stimulation of DNA ligase I. JBiol Chem 277, 1719-1727. 260

263. Thompson, L. H., Brookman, K. W., Jones, N. J., Allen, S. A., and Carrano, A. V. (1990) Molecular cloning of the human XRCC1 gene, which corrects defective DNA strand break repair and sister chromatid exchange. Mol Cell Biol 10, 6160-6171.

264. Cappelli, E., Taylor, R., Cevasco, M., Abbondandolo, A., Caldecott, K., and Frosina, G. (1997) Involvement ofXRCC1 and DNA ligase III gene products in DNA base excision repair. JBiol Chem 272, 23970-23975.

265.Moore, D. J., Taylor, R. M., Clements, P., and Caldecott, K. W. (2000) Mutation of a BRCT domain selectively disrupts DNA single-strand break repair in noncycling Chinese hamster ovary cells. Proc Nati Acad Sci USA 97, 13649-13654.

266. Taylor, R. M., Moore, D. J., Whitehouse, J., Johnson, P., and Caldecott, K. W. (2000) A cell cycle-specific requirement for the XRCC1 BRCT II domain during mammalian DNA strand break repair. Mol Cell Biol 20, 735-740.

267. Tebbs, R. S., Flannery, M. L., Meneses, J. J., Hartmann, A., Tucker, J. D., Thompson, L. H., Cleaver, J. E., and Pedersen, R. A. (1999) Requirement for the Xrccl DNA base excision repair gene during early mouse development. DevBiol 208, 513-529.

268. Tebbs, R. S., Thompson, L. H., and Cleaver, J. E. (2003) Rescue ofXrccl knockout mouse embryo lethality by transgene-complementation. DNA Repair (Amst) 2, 1405-1417.

269.Vidal, A. E., Boiteux, S., Hickson, I. D., and Radicella, J. P. (2001) XRCC1 coordinates the initial and late stages of DNA abasic site repair through protein-protein interactions. EMBO J 20, 6530-6539.

270. Caldecott, K. W., McKeown, C. K., Tucker, J. D., Ljungquist, S., and Thompson, L. H. (1994) An interaction between the mammalian DNA repair protein XRCC1 and DNA ligase III. Mol Cell Biol 14, 68-76.

271. Dianova, II, Sleeth, K. M., Allinson, S. L., Parsons, J. L., Breslin, C., Caldecott, K. W., and Dianov, G. L. (2004) XRCC 1-DNA polymerase beta interaction is required for efficient base excision repair. Nucleic Acids Res 32, 2550-2555. 272. Zhou, J., Ahn, J., Wilson, S. H., and Prives, C. (2001) A role for p53 in base excision repair. EMBO J 20, 914-923. 261

273. Harrigan, J. A., Opresko, P. L., von Kobbe, C., Kedar, P. S., Prasad, R., Wilson, S. H., and Bohr, V. A. (2003) The Werner syndrome protein stimulates DNA polymerase beta strand displacement synthesis via its helicase activity. JBiol Chem 278, 22686-22695.

274.Ahn, B., Harrigan, J. A., Indig, F. B., Wilson, D. M., 3rd, and Bohr, V. A. (2005) Regulation of WRN helicase activity in human base excision repair. J Biol Chem (In press).

275. Li, C., Goodchild, J., and Baril, B. F. (1994) DNA ligase I is associated with the 21 S complex of enzymes for DNA synthesis in HeLa cells. Nucleic Acids Res 22, 632-63 8.

276. Chen, J., Tomkinson, A. E., Ramos, W., Mackey, Z. B., Danehower, S., Walter, C. A., Schultz, R. A., Besterman, J. M., and Husain, I. (1995) Mammalian DNA ligase III: molecular cloning, chromosomal localization, and expression in spermatocytes undergoing meiotic recombination. Mol Cell Biol 15, 54 12-5422.

277. Mackey, Z. B., Ramos, W., Levin, D. S., Walter, C. A., McCarrey, J. R., and Tomkinson, A. B. (1997) An alternative splicing event which occurs in mouse pachytene spennatocytes generates a form of DNA ligase III with distinct biochemical properties that may function in meiotic recombination. Mol Cell Biol 17, 989-998.

278. Wei, Y. F., Robins, P., Carter, K., Caldecott, K., Pappin, D. J., Yu, G. L., Wang, R. P., Shell, B. K., Nash, R. A., and Schar, P. (1995) Molecular cloning and expression of human cDNAs encoding a novel DNA ligase IV and DNA ligase III, an enzyme active in DNA repair and recombination. Mo! Cell Biol 15, 3206-32 16.

279. Nash, R. A., Caldecott, K. W., Barnes, D. E., and Lindahi, T. (1997) XRCC1 protein interacts with one of two distinct forms of DNA ligase III. Biochemistry 36, 5207-5211.

280. Prasad, R., Singhal, R. K., Srivastava, D. K., Molina, J. T., Tomkinson, A. E., and Wilson, S. H. (1996) Specific interaction of DNA polymerase beta and DNA ligase Tin a multiprotein base excision repair complex from bovine testis. JBiol Chem 271, 16000-16007.

281. Dimitriadis, E. K., Prasad, R., Vaske, M. K., Chen, L., Tomkinson, A. B., Lewis, M. S., and Wilson, S. H. (1998) Thermodynamics of human DNA ligase I trimerization and association with DNA polymerase beta. JBiol Chem 273,20540-20550. 262

282. Tom, S., Henricksen, L. A., Park, M. S., and Bambara, R. A. (2001) DNA ligase I and proliferating cell nuclear antigen form a functional complex. JBiol Chem 276,24817-24825.

283. Levin, D. S., McKenna, A. E., Motycka, T. A., Matsumoto, Y., and Tomkinson, A. E. (2000) Interaction between PCNA and DNA ligase I is critical for joining of Okazaki fragments and long-patch base-excision repair. CurrBiol 10, 919-922.

284.Matsumoto, Y., Kim, K., Hurwitz, J., Gary, R., Levin, D. S., Tomkinson, A. E., and Park, M. S. (1999) Reconstitution of proliferating cell nuclear antigen- dependent repair of apurinic/apyrimidinic sites with purified human proteins. J Biol Chem 274, 33703-33708.

285. Burgers, P. M., Koonin, B. V., Bruford, E., Blanco, L., Burtis, K. C., Christman, M. F., Copeland, W. C., Friedberg, E. C., Hanaoka, F., Hinkle, D. C., Lawrence, C. W., Nakanishi, M., Ohmori, H., Prakash, L., Prakash, S., Reynaud, C. A., Sugino, A., Todo, T., Wang, Z., Weill, J. C., and Woodgate, R. (2001) Eukaryotic DNA polymerases: proposal for a revised nomenclature. JB1o1 Chem 276,43487-43490.

286. Miller, I. H. (1972) Experiments in Molecular Genetics, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

287. Marinus, M. G. (1973) Location of DNA methylation genes on the Escherichia co/i K-12 genetic map. Mol Gen Genet 127,47-55.

288. Yanisch-Perron, C., Vieira, J., and Messing, J. Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. 103-119.

289. Bradford, M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248-254.

290. Reisner, A. H. (1984) Gel Protein Stains: A Rapid Procedure. Meth. Enzymol. 104, 439-441.

291. Sambrook, J., Fritsch, E. F., and Maniatis, 1. (1989) Molecular Cloning: A Laboratory Manual, 2nd Ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

292. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the heads of bacteriophage T4. Nature 227, 680-685. 293. Shetlar, M. D., Christensen, J., and Horn, K. (1984) Photochemical addition of amino acids and peptides to DNA. Photochem Photobiol 39, 125-133.

294. Gerland, U., Moroz, J. D., and Hwa, T. (2002) Physical constraints and functional characteristics of transcription factor-DNA interaction. Proc Natl AcadSci USA 99, 12015-12020.

295. Ono, A., and Nishizima, A. (2000) Synthesis of oligodeoxyribonucleotides containing 2'-deoxypseudouridine: inhibition of uracil-DNA glycosylase. Nucleic Acids Symp Ser, 127-128.

296. Nordlund, T. M., Andersson, S., Nilsson, L., Rigler, R., Graslund, A., and McLaughlin, L. W. (1989) Structure and dynamics of a fluorescent DNA oligomer containing the EcoRI recognition sequence: fluorescence, molecular dynamics, and NMR studies. Biochemistry 28, 9095-9 103.

297. Stivers, J. T. (1998) 2-Aminopurine fluorescence studies of base stacking interactions at abasic sites in DNA: metal-ion and base sequence effects. Nucleic Acids Res 26, 3837-3 844.

298. Jean, J. M., and Hall, K. B. (2001) 2-Aminopurine fluorescence quenching and lifetimes: role of base stacking. Proc NatlAcadSci USA 98, 37-41.

299. Dunlap, C. A., and Tsai, M. D. (2002) Use of 2-aminopurine and tryptophan fluorescence as probes in kinetic analyses of DNA polymerase beta. Biochemistry 41, 11226-11235.

300.Rai, P., Cole, T. D., Thompson, E., Millar, D. P., and Linn, S. (2003) Steady- state and time-resolved fluorescence studies indicate an unusual conformation of 2-aminopurine within ATAT and TATA duplex DNA sequences. Nucleic Acids Res 31, 2323-2332.

301. Ward, D. C., Reich, E., and Stryer, L. (1969) Fluorescence studies of nucleotides and polynucleotides. I. Formycin, 2-aminopurine riboside, 2,6- diaminopurine riboside, and their derivatives. JBiol Chem 244, 1228-1237.

302. Lundquist, A. J. (1999) Mechanism of action of Escherichia coli uracil-DNA glycosylase and interaction with the bacteriophage PB S-2 uracil-DNA glycosylase inhibitor protein. Ph.D. thesis, Oregon State University, Corvallis, OR

303. Jiang, Y. L., Kwon, K., and Stivers, J. T. (2001) Turning On uracil-DNA glycosylase using a pyrene nucleotide switch. JBiol Chem 276,42347-42354.